|
{ |
|
"metadata": { |
|
"ParamSize": 405, |
|
"ParamBytes": 8259194880.0, |
|
"BitsPerParam": 5.0008474901387725 |
|
}, |
|
"records": [ |
|
{ |
|
"dataPath": "params_shard_0.bin", |
|
"format": "raw-shard", |
|
"nbytes": 131072000, |
|
"records": [ |
|
{ |
|
"name": "lm_head.q_weight", |
|
"shape": [ |
|
51200, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 131072000, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6810a3452a28ff93431ae937e5d3004c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_1.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3f294b3a146447848d9ec7d5572652c1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_2.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "00df18affbd1d4793e0400896e76d254" |
|
}, |
|
{ |
|
"dataPath": "params_shard_3.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "603c9dd85a47f925f0703dcc0de63faa" |
|
}, |
|
{ |
|
"dataPath": "params_shard_4.bin", |
|
"format": "raw-shard", |
|
"nbytes": 29685760, |
|
"records": [ |
|
{ |
|
"name": "lm_head.q_scale", |
|
"shape": [ |
|
51200, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 16384000, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.30.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 16384000 |
|
}, |
|
{ |
|
"name": "model.layers.30.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 16394240 |
|
}, |
|
{ |
|
"name": "model.layers.30.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 20817920 |
|
}, |
|
{ |
|
"name": "model.layers.30.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 29665280 |
|
}, |
|
{ |
|
"name": "model.layers.31.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 29675520 |
|
} |
|
], |
|
"md5sum": "19820819fa2f265b4221727c629a7c7d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_5.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e8d243a865eca76f5ebc331bd3232f41" |
|
}, |
|
{ |
|
"dataPath": "params_shard_6.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f73a1afaf82ccccb43dc12133db3724f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_7.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "81067ebb4ab47571af08298fd268b48e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_8.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.31.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.31.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.31.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 18196480 |
|
}, |
|
{ |
|
"name": "model.layers.31.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 31303680 |
|
}, |
|
{ |
|
"name": "model.layers.32.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "3f3a36fcb0ee1b8fee117563370b73f6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_9.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b2f3baaf95dc86a5de9e7d22c6892943" |
|
}, |
|
{ |
|
"dataPath": "params_shard_10.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fe1eab67a9c8c3dc212de6a78a1aa959" |
|
}, |
|
{ |
|
"dataPath": "params_shard_11.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fe632bc96a74131618bda4e69fd4933e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_12.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.32.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.32.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.32.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 18196480 |
|
}, |
|
{ |
|
"name": "model.layers.32.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 31303680 |
|
}, |
|
{ |
|
"name": "model.layers.33.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "913db4755962573b8f36aa13d36bc0d4" |
|
}, |
|
{ |
|
"dataPath": "params_shard_13.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e17b4697b0f10bc728ba562ef55e2764" |
|
}, |
|
{ |
|
"dataPath": "params_shard_14.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "42b96e1d6801b0cddc84bebeecbefa05" |
|
}, |
|
{ |
|
"dataPath": "params_shard_15.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "63da875dc0bd10ef70e75eb337f5076b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_16.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.33.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.33.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.33.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 18196480 |
|
}, |
|
{ |
|
"name": "model.layers.33.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 31303680 |
|
}, |
|
{ |
|
"name": "model.layers.34.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "d95d639133bcd0cb07a0877fab5648c6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_17.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "8b6fc13f53568ae79d660dfe9f9f3433" |
|
}, |
|
{ |
|
"dataPath": "params_shard_18.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e55e05c71cba1e1ec900b27ee5482d5d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_19.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9dd8b6afd4b4cbc530ffbc80165b712a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_20.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.34.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.34.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.34.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 18196480 |
|
}, |
|
{ |
|
"name": "model.layers.34.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 31303680 |
|
}, |
|
{ |
|
"name": "model.layers.35.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "2b858b8e5354b5a675fb3df509c4365b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_21.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "896257ac759a8b88cd6fb3c754b4e1e8" |
|
}, |
|
{ |
|
"dataPath": "params_shard_22.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "195f87d72cf2823f3a03d70c71b643ff" |
|
}, |
|
{ |
|
"dataPath": "params_shard_23.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d9af87c521c8ed15486541e08a508900" |
|
}, |
|
{ |
|
"dataPath": "params_shard_24.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.35.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.35.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.35.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 18196480 |
|
}, |
|
{ |
|
"name": "model.layers.35.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 31303680 |
|
}, |
|
{ |
|
"name": "model.layers.36.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "e22ab163c3697ef17a709e776954a927" |
|
}, |
|
{ |
|
"dataPath": "params_shard_25.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "58eef848b20c76fed58f606e2f2a9fed" |
|
}, |
|
{ |
|
"dataPath": "params_shard_26.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "bdf6c37cdb66478842f4c1b26a4305ca" |
|
}, |
|
{ |
|
"dataPath": "params_shard_27.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "81642b7389146762900ea28fc464bf2a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_28.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.36.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.36.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.36.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 18196480 |
|
}, |
|
{ |
|
"name": "model.layers.36.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 31303680 |
|
}, |
|
{ |
|
"name": "model.layers.37.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "b48db3dd9f7faa34a81b63593f4806b0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_29.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d6200406637c7f720f6011465b42e8c0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_30.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "89cdd7660202b9ec44a5d9398f7d71c1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_31.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e7e15dc1920e7d21bf2a44fbef4a1d77" |
|
}, |
|
{ |
|
"dataPath": "params_shard_32.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.37.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.37.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.37.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 18196480 |
|
}, |
|
{ |
|
"name": "model.layers.37.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 31303680 |
|
}, |
|
{ |
|
"name": "model.layers.38.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "588324e8bf059ccae6c590fda9a27fe8" |
|
}, |
|
{ |
|
"dataPath": "params_shard_33.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f25a7aaf529a97ca3e7b156de90fc4ef" |
|
}, |
|
{ |
|
"dataPath": "params_shard_34.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "87b549322480e05200b9977c582c2c8f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_35.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b4fdde3edc5101abafeae00769309de3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_36.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.38.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.38.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.38.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 18196480 |
|
}, |
|
{ |
|
"name": "model.layers.38.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 31303680 |
|
}, |
|
{ |
|
"name": "model.layers.39.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "ddd26e3455d1cecc4cc0f37cec242f86" |
|
}, |
|
{ |
|
"dataPath": "params_shard_37.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e54b8753ff92f88a57f6a62ecab40145" |
|
}, |
|
{ |
|
"dataPath": "params_shard_38.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fac0334d0427a21535a1649ecdc7bc94" |
|
}, |
|
{ |
|
"dataPath": "params_shard_39.bin", |
|
"format": "raw-shard", |
|
"nbytes": 131072000, |
|
"records": [ |
|
{ |
|
"name": "model.embed_tokens.q_weight", |
|
"shape": [ |
|
51200, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 131072000, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "0c0300e527d04acaa36f3798381ace77" |
|
}, |
|
{ |
|
"dataPath": "params_shard_40.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.39.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.39.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 4423680 |
|
}, |
|
{ |
|
"name": "model.layers.39.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 13271040 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 13281280 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 18196480 |
|
}, |
|
{ |
|
"name": "model.layers.39.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 31303680 |
|
}, |
|
{ |
|
"name": "model.norm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "1252dc41a0cd35e02317a5715bf37ba0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_41.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "36b687f5f19894c1605c3d2a30d202ca" |
|
}, |
|
{ |
|
"dataPath": "params_shard_42.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "81983a6ea328170a32801ec9a88b730a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_43.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9c014c065724d4ca3916144663f1cc34" |
|
}, |
|
{ |
|
"dataPath": "params_shard_44.bin", |
|
"format": "raw-shard", |
|
"nbytes": 29675520, |
|
"records": [ |
|
{ |
|
"name": "model.embed_tokens.q_scale", |
|
"shape": [ |
|
51200, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 16384000, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.0.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 16384000 |
|
}, |
|
{ |
|
"name": "model.layers.0.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 16394240 |
|
}, |
|
{ |
|
"name": "model.layers.0.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 20817920 |
|
}, |
|
{ |
|
"name": "model.layers.0.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 29665280 |
|
} |
|
], |
|
"md5sum": "997ad33e4f9b9fa11ddbea60064e9fd3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_45.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c0ae0eea0a8ef38ba45aa5416acb2579" |
|
}, |
|
{ |
|
"dataPath": "params_shard_46.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "edac1b63c0a064a5c2c0570766d2c64d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_47.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b7f4c682192d0819d4812d577681b06a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_48.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.0.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.0.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.1.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.1.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.1.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.1.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "124caff11eead1c638370f269b33f83b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_49.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9b27921af5f1e99b512e6c7a6abbd574" |
|
}, |
|
{ |
|
"dataPath": "params_shard_50.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1144fe42380536d9bdad2190b32f3fb3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_51.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1f5bd123762d2f13c32afb39056215da" |
|
}, |
|
{ |
|
"dataPath": "params_shard_52.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.1.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.1.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.10.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.10.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.10.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.10.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "c88a57a5fafcd9f314893a16a5ea014e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_53.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e305208a1c3626cc188d5fb08b37326d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_54.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "9523979cd321a2d169e4df4daaeeeb45" |
|
}, |
|
{ |
|
"dataPath": "params_shard_55.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a6d7884a727ecd0952b65bf7dcdc7fca" |
|
}, |
|
{ |
|
"dataPath": "params_shard_56.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.10.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.10.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.11.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.11.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.11.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.11.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "3c9fa1656b6d448b999c7108c14fbf9e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_57.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e7c3a9af2c537f0b61ea2637604b4e47" |
|
}, |
|
{ |
|
"dataPath": "params_shard_58.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "39f2f5b20b3afd9b4d8cfbf50e600ecb" |
|
}, |
|
{ |
|
"dataPath": "params_shard_59.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "743d573baeb05622ad5ae7161bd1d2cc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_60.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.11.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.11.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.12.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.12.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.12.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.12.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "45dc05f067f5058ff9a001ac198956e3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_61.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "83b96e7f3d6bcb5456e95917d54fa816" |
|
}, |
|
{ |
|
"dataPath": "params_shard_62.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "72850d44d3b2d1267e1ed507bf6a1234" |
|
}, |
|
{ |
|
"dataPath": "params_shard_63.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d425d33801edf3a918f1159dbea35e05" |
|
}, |
|
{ |
|
"dataPath": "params_shard_64.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.12.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.12.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.13.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.13.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.13.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.13.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "2e53c246415fcff9077c0ffb839e2ff1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_65.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6700c1ef39d4989ce2293cc778c60694" |
|
}, |
|
{ |
|
"dataPath": "params_shard_66.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "cc09df75fa8e47205b7d5a1e6b98f8b5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_67.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6a3a08be4b6960a0620ec42437689c11" |
|
}, |
|
{ |
|
"dataPath": "params_shard_68.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32931840, |
|
"records": [ |
|
{ |
|
"name": "model.layers.13.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.13.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.14.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.14.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24084480 |
|
} |
|
], |
|
"md5sum": "e06c6eb2183af88545769fc3bcbffcca" |
|
}, |
|
{ |
|
"dataPath": "params_shard_69.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "95d86d32e5341b2b80ff13c465215d57" |
|
}, |
|
{ |
|
"dataPath": "params_shard_70.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "508c5587f30f802242767812b53ea47d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_71.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fc02240d67da2b4b19aac35270b22c58" |
|
}, |
|
{ |
|
"dataPath": "params_shard_72.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.14.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.14.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.2.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.2.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.2.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.2.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "72411756c27fd97a0d692a02af01778a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_73.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "ffc5113ce353433b6c5b12a197a4f560" |
|
}, |
|
{ |
|
"dataPath": "params_shard_74.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "8cfa400a362f7bb50f9c56e05acc20cc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_75.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "96c1f10bf1155b710886deb3f3eeef38" |
|
}, |
|
{ |
|
"dataPath": "params_shard_76.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.2.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.2.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.3.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.3.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.3.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.3.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "4ff0f2b49419ac38b2086094b861b17c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_77.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "af0e3fa2b66c7c4b92e77676bab72911" |
|
}, |
|
{ |
|
"dataPath": "params_shard_78.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b59bf9b9d455d0bc4d3607a71b7a2106" |
|
}, |
|
{ |
|
"dataPath": "params_shard_79.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "009f1c6391d01056d62f2b3d9ff5d156" |
|
}, |
|
{ |
|
"dataPath": "params_shard_80.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.3.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.3.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.4.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.4.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.4.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.4.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "6e50ee207223124ad08355316211df64" |
|
}, |
|
{ |
|
"dataPath": "params_shard_81.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "272e65feef00aaee8b121270812e1147" |
|
}, |
|
{ |
|
"dataPath": "params_shard_82.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "3cb4a6cd55ff828f5b2894031802f61e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_83.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f5f90c092bd77dfe44559aa2df5a0ac6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_84.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.4.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.4.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.5.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.5.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.5.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.5.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "12c6da793959157e3d322a962b9a035f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_85.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6064c32ce282d98fa35e6d32de78f6d1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_86.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "50a2364ab3834209b7f518c4c8426ae3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_87.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c9466e736aab370392c9f9caee32edd3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_88.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.5.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.5.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.6.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.6.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.6.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.6.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "42a2fbf38962061a1c11aa57ba577fae" |
|
}, |
|
{ |
|
"dataPath": "params_shard_89.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "00611e3e8f6ccfe0bc5f5694dcfb6f9f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_90.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2614afe64ab43a198027ab26a64aa685" |
|
}, |
|
{ |
|
"dataPath": "params_shard_91.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "98cf30cb3706180b5e74177848fea2e2" |
|
}, |
|
{ |
|
"dataPath": "params_shard_92.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.6.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.6.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.7.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.7.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.7.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.7.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "fcb01bd73e6b863e28484963b2e10aed" |
|
}, |
|
{ |
|
"dataPath": "params_shard_93.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a6336ee8ac08fb95761f16c0923dd607" |
|
}, |
|
{ |
|
"dataPath": "params_shard_94.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "d0202e51e818ca75af6f044daca18c39" |
|
}, |
|
{ |
|
"dataPath": "params_shard_95.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "09a766f09ac44b8ef71e7ef7273d6c46" |
|
}, |
|
{ |
|
"dataPath": "params_shard_96.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.7.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.7.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.8.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.8.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.8.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.8.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "66006be6aff7b5d6cac50d0f294e7b6f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_97.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "cee47a680d58dc25617f08ab08c5f74c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_98.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a0f6ecead3b4b5ddbf5972979ff3d626" |
|
}, |
|
{ |
|
"dataPath": "params_shard_99.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "be20cfc3012f1d89580b99d3d04d335b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_100.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.8.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.8.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.9.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.9.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.9.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.9.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "0b1db6dfef578af164b79b34f78dea33" |
|
}, |
|
{ |
|
"dataPath": "params_shard_101.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "63317074e615201e142d707facdf4c5c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_102.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c661ea4f1e34c1708a27f1dd01befeff" |
|
}, |
|
{ |
|
"dataPath": "params_shard_103.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "97c0db4500ffc5612b3b44334ee08422" |
|
}, |
|
{ |
|
"dataPath": "params_shard_104.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32972800, |
|
"records": [ |
|
{ |
|
"name": "model.layers.9.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.9.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.14.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.14.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.15.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19681280 |
|
}, |
|
{ |
|
"name": "model.layers.15.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19691520 |
|
}, |
|
{ |
|
"name": "model.layers.15.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24115200 |
|
}, |
|
{ |
|
"name": "model.layers.15.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32962560 |
|
} |
|
], |
|
"md5sum": "8f445ec09de11580efeabb12a3e4ef54" |
|
}, |
|
{ |
|
"dataPath": "params_shard_105.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "ec18c6f694f46b6c46ed1559b4cfb007" |
|
}, |
|
{ |
|
"dataPath": "params_shard_106.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "689bee8cf17878ecfbde86bb974e0a40" |
|
}, |
|
{ |
|
"dataPath": "params_shard_107.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "dbad7d8d212db3822120e6460808d4ca" |
|
}, |
|
{ |
|
"dataPath": "params_shard_108.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.15.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.15.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.16.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.16.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.16.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.16.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "a6d0f9e2a242cceaf41a7b30cc18f4c5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_109.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "09ffd4628b7b1bcd79374d8f1dde1c01" |
|
}, |
|
{ |
|
"dataPath": "params_shard_110.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "98caf3ec7c59ed9cd24a38edb1f47332" |
|
}, |
|
{ |
|
"dataPath": "params_shard_111.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "abc1e3e4b97b2725f2a9a9173387745d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_112.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.16.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.16.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.17.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.17.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.17.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.17.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "57bd3dd968f51ada5bf137e64afaf60b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_113.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c6bbbc6949d6d4441bad61b07b622563" |
|
}, |
|
{ |
|
"dataPath": "params_shard_114.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "6de2c548fa716c6149f94e3f9d7653a5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_115.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "90edff7b42f96cef68e93abf919bba44" |
|
}, |
|
{ |
|
"dataPath": "params_shard_116.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.17.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.17.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.18.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.18.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.18.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.18.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "3723ffcc60adcd565ca174e476beada1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_117.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "90e5c099757e3c00973e2417aa2bce0d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_118.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a3c65a9405cf6a4ef644141c14d5b86c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_119.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "730d90d5233ebe74091d756d84318368" |
|
}, |
|
{ |
|
"dataPath": "params_shard_120.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.18.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.18.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.19.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.19.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.19.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.19.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "e57bda9a22b5f7eb7d461d85d22c8eef" |
|
}, |
|
{ |
|
"dataPath": "params_shard_121.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "67232676702577c7e54588313f3212d5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_122.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "e6699b29795333e4829a83e7a73d9ca5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_123.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "76057dfeee856af8826d9701f2ed1553" |
|
}, |
|
{ |
|
"dataPath": "params_shard_124.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.19.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.19.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.20.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.20.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.20.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.20.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "adac4ff44c812d7db7646a8aebdb57fa" |
|
}, |
|
{ |
|
"dataPath": "params_shard_125.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1123ac8ace40bddffbe9e04e7c8af654" |
|
}, |
|
{ |
|
"dataPath": "params_shard_126.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fdd6a6beeead06e1f8ff20b5e5748f80" |
|
}, |
|
{ |
|
"dataPath": "params_shard_127.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "8161f387ba0d8e8dc310d1ec41411b53" |
|
}, |
|
{ |
|
"dataPath": "params_shard_128.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.20.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.20.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.21.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.21.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.21.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.21.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "571fa26258b553cd1203d6108c00bcbd" |
|
}, |
|
{ |
|
"dataPath": "params_shard_129.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "8109c9c217a699033c54ba60c696ef0d" |
|
}, |
|
{ |
|
"dataPath": "params_shard_130.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "fee76ffa8113594655d24b2943b6e6c1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_131.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f4ed675dffcfca18d74dad3155fb55a3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_132.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.21.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.21.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.22.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.22.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.22.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.22.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "9de7b4b50cb2a47f2d1e69fbb68f46b1" |
|
}, |
|
{ |
|
"dataPath": "params_shard_133.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "79590e16bf56f075029963fe7c4ac0c7" |
|
}, |
|
{ |
|
"dataPath": "params_shard_134.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "521539811886e365f77de99b1103b446" |
|
}, |
|
{ |
|
"dataPath": "params_shard_135.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "31a3521bd8ac93431acd41ebdd8a87cc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_136.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.22.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.22.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.23.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.23.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.23.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.23.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "bce7ea7ee57b82860b83e02078990dba" |
|
}, |
|
{ |
|
"dataPath": "params_shard_137.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f15a7bc1928fb5742fb641b7534c8e1b" |
|
}, |
|
{ |
|
"dataPath": "params_shard_138.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "f93bd3b7c9ece4de7d8cf3df2f4258a8" |
|
}, |
|
{ |
|
"dataPath": "params_shard_139.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "66f38206a6bf1c532d7c7e3b222f9b45" |
|
}, |
|
{ |
|
"dataPath": "params_shard_140.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.23.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.23.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.24.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.24.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.24.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.24.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "39886c4e039ae814d4291d41d6425ae5" |
|
}, |
|
{ |
|
"dataPath": "params_shard_141.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "4e43c6b2b3d9eed3f893a6644e65f925" |
|
}, |
|
{ |
|
"dataPath": "params_shard_142.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "bd987b429034876a41be6669f9dc5be9" |
|
}, |
|
{ |
|
"dataPath": "params_shard_143.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2fa595cbd3dee3ec901e2d56662f79b0" |
|
}, |
|
{ |
|
"dataPath": "params_shard_144.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.24.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.24.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.25.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.25.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.25.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.25.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "95717d789d6a345c5d4cc94e44ab7100" |
|
}, |
|
{ |
|
"dataPath": "params_shard_145.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "2e9be658297d71990935277a4d3f0358" |
|
}, |
|
{ |
|
"dataPath": "params_shard_146.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "c6c6b0fe93976a81a00a428386b11053" |
|
}, |
|
{ |
|
"dataPath": "params_shard_147.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b00681ba40a65ec9b06562621376654f" |
|
}, |
|
{ |
|
"dataPath": "params_shard_148.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.25.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.25.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.26.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.26.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.26.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.26.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "54fd242b08157656fdf2fe02cd636b5e" |
|
}, |
|
{ |
|
"dataPath": "params_shard_149.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "89a78dbc223bb600632a2c6ca399f6ff" |
|
}, |
|
{ |
|
"dataPath": "params_shard_150.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b50f49f811a2c3f77cb392b7f227fd1a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_151.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "1bb8581320593d960c597c676c6ec3da" |
|
}, |
|
{ |
|
"dataPath": "params_shard_152.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.26.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.26.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.27.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.27.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.27.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.27.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "729990f2a4ca2d86c8a76b19738fcf8a" |
|
}, |
|
{ |
|
"dataPath": "params_shard_153.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "a27d6be776a0fc87543ef4471619dee3" |
|
}, |
|
{ |
|
"dataPath": "params_shard_154.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "81d37ee565f62960a9eeef70410cbd73" |
|
}, |
|
{ |
|
"dataPath": "params_shard_155.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "b08e3f9991e09f6592e97a4f040050fc" |
|
}, |
|
{ |
|
"dataPath": "params_shard_156.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.27.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.27.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.28.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.28.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.28.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.28.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "eb13d53aea48c4b0d0a942cd60039a41" |
|
}, |
|
{ |
|
"dataPath": "params_shard_157.bin", |
|
"format": "raw-shard", |
|
"nbytes": 35389440, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.mlp.down_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
1728 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 35389440, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "5f143bc632f4f454e5fac6da0f703c35" |
|
}, |
|
{ |
|
"dataPath": "params_shard_158.bin", |
|
"format": "raw-shard", |
|
"nbytes": 70778880, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.mlp.gate_up_proj.q_weight", |
|
"shape": [ |
|
27648, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 70778880, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "07a36d239ea960e316f14ea7760955b6" |
|
}, |
|
{ |
|
"dataPath": "params_shard_159.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "be1285af1784ca4e337a0043bf636939" |
|
}, |
|
{ |
|
"dataPath": "params_shard_160.bin", |
|
"format": "raw-shard", |
|
"nbytes": 32952320, |
|
"records": [ |
|
{ |
|
"name": "model.layers.28.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.28.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.28.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.29.input_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 19660800 |
|
}, |
|
{ |
|
"name": "model.layers.29.mlp.down_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
432 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4423680, |
|
"byteOffset": 19671040 |
|
}, |
|
{ |
|
"name": "model.layers.29.mlp.gate_up_proj.q_scale", |
|
"shape": [ |
|
27648, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 8847360, |
|
"byteOffset": 24094720 |
|
}, |
|
{ |
|
"name": "model.layers.29.post_attention_layernorm.weight", |
|
"shape": [ |
|
5120 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 10240, |
|
"byteOffset": 32942080 |
|
} |
|
], |
|
"md5sum": "fff685a92624db67343001f50c47782c" |
|
}, |
|
{ |
|
"dataPath": "params_shard_161.bin", |
|
"format": "raw-shard", |
|
"nbytes": 39321600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.self_attn.qkv_proj.q_weight", |
|
"shape": [ |
|
15360, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 39321600, |
|
"byteOffset": 0 |
|
} |
|
], |
|
"md5sum": "546a8c5a28ac28b13352a7a5205a6d68" |
|
}, |
|
{ |
|
"dataPath": "params_shard_162.bin", |
|
"format": "raw-shard", |
|
"nbytes": 24576000, |
|
"records": [ |
|
{ |
|
"name": "model.layers.29.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.29.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 4915200 |
|
}, |
|
{ |
|
"name": "model.layers.29.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 18022400 |
|
}, |
|
{ |
|
"name": "model.layers.30.self_attn.qkv_proj.q_scale", |
|
"shape": [ |
|
15360, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 4915200, |
|
"byteOffset": 19660800 |
|
} |
|
], |
|
"md5sum": "7229a6e53515a622389dd4a94f3d42ec" |
|
}, |
|
{ |
|
"dataPath": "params_shard_163.bin", |
|
"format": "raw-shard", |
|
"nbytes": 14745600, |
|
"records": [ |
|
{ |
|
"name": "model.layers.30.self_attn.o_proj.q_weight", |
|
"shape": [ |
|
5120, |
|
640 |
|
], |
|
"dtype": "uint32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 13107200, |
|
"byteOffset": 0 |
|
}, |
|
{ |
|
"name": "model.layers.30.self_attn.o_proj.q_scale", |
|
"shape": [ |
|
5120, |
|
160 |
|
], |
|
"dtype": "float32", |
|
"format": "f32-to-bf16", |
|
"nbytes": 1638400, |
|
"byteOffset": 13107200 |
|
} |
|
], |
|
"md5sum": "64cfba1d1340a7299beaed0894ed3ed7" |
|
} |
|
] |
|
} |