dacorvo HF Staff commited on
Commit
5a251a1
·
verified ·
1 Parent(s): c2ec2c0

Synchronizing local compiler cache.

Browse files
Files changed (17) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0.dev7/llama/meta-llama/Meta-Llama-3.1-8B/bccd957d3ee1497242ee.json +77 -0
  3. neuronxcc-2.17.194.0+d312836f/MODULE_13d43d0be39ff7b1d66f+bfe5714b/compile_flags.json +1 -0
  4. neuronxcc-2.17.194.0+d312836f/MODULE_13d43d0be39ff7b1d66f+bfe5714b/model.done +0 -0
  5. neuronxcc-2.17.194.0+d312836f/MODULE_13d43d0be39ff7b1d66f+bfe5714b/model.hlo_module.pb +3 -0
  6. neuronxcc-2.17.194.0+d312836f/MODULE_13d43d0be39ff7b1d66f+bfe5714b/model.neff +3 -0
  7. neuronxcc-2.17.194.0+d312836f/MODULE_5dc7b4d69bf22af2563b+165e9558/model.hlo_module.pb +1 -1
  8. neuronxcc-2.17.194.0+d312836f/MODULE_5dc7b4d69bf22af2563b+165e9558/model.neff +1 -1
  9. neuronxcc-2.17.194.0+d312836f/MODULE_5dc7b4d69bf22af2563b+165e9558/wrapped_neff.hlo +1 -1
  10. neuronxcc-2.17.194.0+d312836f/MODULE_6a90845afeaebacf0657+165e9558/model.hlo_module.pb +1 -1
  11. neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/compile_flags.json +1 -0
  12. neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/model.done +0 -0
  13. neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/model.hlo_module.pb +3 -0
  14. neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/model.neff +3 -0
  15. neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/wrapped_neff.hlo +3 -0
  16. neuronxcc-2.17.194.0+d312836f/MODULE_f4c8e9585d2c42f74ede+bfe5714b/model.hlo_module.pb +1 -1
  17. neuronxcc-2.17.194.0+d312836f/MODULE_f4c8e9585d2c42f74ede+bfe5714b/model.neff +1 -1
.gitattributes CHANGED
@@ -8327,3 +8327,6 @@ neuronxcc-2.17.194.0+d312836f/MODULE_c22409dc369cb240afe5+431f5505/model.neff fi
8327
  neuronxcc-2.17.194.0+d312836f/MODULE_5dc7b4d69bf22af2563b+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
8328
  neuronxcc-2.17.194.0+d312836f/MODULE_5dc7b4d69bf22af2563b+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
8329
  neuronxcc-2.17.194.0+d312836f/MODULE_f4c8e9585d2c42f74ede+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
8327
  neuronxcc-2.17.194.0+d312836f/MODULE_5dc7b4d69bf22af2563b+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
8328
  neuronxcc-2.17.194.0+d312836f/MODULE_5dc7b4d69bf22af2563b+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
8329
  neuronxcc-2.17.194.0+d312836f/MODULE_f4c8e9585d2c42f74ede+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
8330
+ neuronxcc-2.17.194.0+d312836f/MODULE_13d43d0be39ff7b1d66f+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
8331
+ neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
8332
+ neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0.dev7/llama/meta-llama/Meta-Llama-3.1-8B/bccd957d3ee1497242ee.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "meta-llama/Meta-Llama-3.1-8B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 4096,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 14336,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 8,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "meta-llama/Meta-Llama-3.1-8B",
26
+ "checkpoint_revision": "d04e592bb4f6aa9cfee91e2e20afa771667e1d4b",
27
+ "continuous_batching": true,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 8,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 8,
37
+ "max_context_length": 4096,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 4096,
42
+ "neuronxcc_version": "2.17.194.0+d312836f",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": true,
45
+ "optimum_neuron_version": "0.2.0.dev7",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "bfloat16",
52
+ "sequence_length": 4096,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "bfloat16",
58
+ "tp_degree": 8,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 32,
62
+ "num_hidden_layers": 32,
63
+ "num_key_value_heads": 8,
64
+ "pretraining_tp": 1,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_scaling": {
67
+ "factor": 8.0,
68
+ "high_freq_factor": 4.0,
69
+ "low_freq_factor": 1.0,
70
+ "original_max_position_embeddings": 8192,
71
+ "rope_type": "llama3"
72
+ },
73
+ "rope_theta": 500000.0,
74
+ "tie_word_embeddings": false,
75
+ "use_cache": true,
76
+ "vocab_size": 128256
77
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_13d43d0be39ff7b1d66f+bfe5714b/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
neuronxcc-2.17.194.0+d312836f/MODULE_13d43d0be39ff7b1d66f+bfe5714b/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_13d43d0be39ff7b1d66f+bfe5714b/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:419739a2aed66ac70ddb7df7bb17ae7e2ac550d3db912d2e8eebade1854aa494
3
+ size 919863
neuronxcc-2.17.194.0+d312836f/MODULE_13d43d0be39ff7b1d66f+bfe5714b/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd861cf17e56f6087cda7789042339a8c44ca197e57f4c4282c5203df76f6107
3
+ size 32646144
neuronxcc-2.17.194.0+d312836f/MODULE_5dc7b4d69bf22af2563b+165e9558/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9ad1f4ff795798cade75282ae94a7ac55f76202dc01647366eafa41bfc21cc9b
3
  size 776785
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05e86645a58331932d66831baddd9b7e98aedcd00e5d6a6ba3ca9ee748a0eee5
3
  size 776785
neuronxcc-2.17.194.0+d312836f/MODULE_5dc7b4d69bf22af2563b+165e9558/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:51cbd827c200db4053c928cbfda9df76574cd38f07753405bdf8402e844e8fcf
3
  size 4619264
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72ec43433e1f13bf26c7d715b18d338f3479f9c04557eb266b92daac38057ef7
3
  size 4619264
neuronxcc-2.17.194.0+d312836f/MODULE_5dc7b4d69bf22af2563b+165e9558/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:98bb3c5284bfca0e61ec7f73a869320ae27c5ecb4e25f81913b4b346403a184e
3
  size 4757208
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b628d58b4f138f2cbc3738a81e3a4a0373c4020b5fa4e34d63f0f21403cd0bbb
3
  size 4757208
neuronxcc-2.17.194.0+d312836f/MODULE_6a90845afeaebacf0657+165e9558/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9a248564bc8d77e6620b1bedf41885c8476b5052a11425ae8c2d2a2c41edbc58
3
  size 756302
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b248364d3a2a5da37f408422d71899f7e614f25f55f9d56a85ee52656c462eb
3
  size 756302
neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt --enable-internal-neff-wrapper"
neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eff4fd589b7cfffc5a2012d4ff821587a751e7802d3309d758063ea31a559388
3
+ size 776785
neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eddfd930ab2a4960ae9a82ef4777532bfe4b642f32da5aaff389576713c271a9
3
+ size 5192704
neuronxcc-2.17.194.0+d312836f/MODULE_f303673a5c27cd35d2e8+165e9558/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf5ffd60d34a030abe0fd072116cdde45caf2be6be565269b737d30c5825f0fe
3
+ size 5330648
neuronxcc-2.17.194.0+d312836f/MODULE_f4c8e9585d2c42f74ede+bfe5714b/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:514435fea3eb282a99100018f5e7df88beca2bc96c0231a502cd289bd871301a
3
  size 919863
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cce349ddee242bc28cfe0074d257ff0f05e3071140d3725bc47d17e75daeccc5
3
  size 919863
neuronxcc-2.17.194.0+d312836f/MODULE_f4c8e9585d2c42f74ede+bfe5714b/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ef04be1a5382cc99d58226d5a88981bc93d29da18f62e79529a3a3892843c62c
3
  size 32646144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b8ce099e2add20a334a5c35af63e213f7e44360334aeb770ae8bd1abb6962f8
3
  size 32646144