Synchronizing local compiler cache.
Browse files- .gitattributes +2 -0
- neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0.dev2/inference/granite/ibm-granite/granite-3.1-2b-instruct/245ec3826de5d0d8c4f3.json +41 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_5b7940618fd1ad4a8869+613edded/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_5b7940618fd1ad4a8869+613edded/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_5b7940618fd1ad4a8869+613edded/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_5b7940618fd1ad4a8869+613edded/model.neff +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_9f5a907ba7db9a6a1e51+613edded/compile_flags.json +1 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_9f5a907ba7db9a6a1e51+613edded/model.done +0 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_9f5a907ba7db9a6a1e51+613edded/model.hlo_module.pb +3 -0
- neuronxcc-2.17.194.0+d312836f/MODULE_9f5a907ba7db9a6a1e51+613edded/model.neff +3 -0
.gitattributes
CHANGED
@@ -7657,3 +7657,5 @@ neuronxcc-2.17.194.0+d312836f/MODULE_eb1b42c1b8d27e328577+613edded/model.neff fi
|
|
7657 |
neuronxcc-2.17.194.0+d312836f/MODULE_f6243540544ea23fcfe1+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7658 |
neuronxcc-2.17.194.0+d312836f/MODULE_f67a180204d2661c3ced+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7659 |
neuronxcc-2.17.194.0+d312836f/MODULE_fac5455884c9dfdee723+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
7657 |
neuronxcc-2.17.194.0+d312836f/MODULE_f6243540544ea23fcfe1+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7658 |
neuronxcc-2.17.194.0+d312836f/MODULE_f67a180204d2661c3ced+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7659 |
neuronxcc-2.17.194.0+d312836f/MODULE_fac5455884c9dfdee723+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7660 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_5b7940618fd1ad4a8869+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
7661 |
+
neuronxcc-2.17.194.0+d312836f/MODULE_9f5a907ba7db9a6a1e51+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0.dev2/inference/granite/ibm-granite/granite-3.1-2b-instruct/245ec3826de5d0d8c4f3.json
ADDED
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_entry_class": "SingleModelCacheEntry",
|
3 |
+
"_model_id": "ibm-granite/granite-3.1-2b-instruct",
|
4 |
+
"_task": "text-generation",
|
5 |
+
"architectures": [
|
6 |
+
"GraniteForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_bias": false,
|
9 |
+
"attention_dropout": 0.1,
|
10 |
+
"attention_multiplier": 0.015625,
|
11 |
+
"embedding_multiplier": 12.0,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 2048,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 8192,
|
16 |
+
"logits_scaling": 8.0,
|
17 |
+
"max_position_embeddings": 131072,
|
18 |
+
"mlp_bias": false,
|
19 |
+
"model_type": "granite",
|
20 |
+
"neuron": {
|
21 |
+
"auto_cast_type": "bf16",
|
22 |
+
"batch_size": 8,
|
23 |
+
"checkpoint_id": "ibm-granite/granite-3.1-2b-instruct",
|
24 |
+
"checkpoint_revision": "374ef54e020a3ce208c65e96d6213922a87d8952",
|
25 |
+
"compiler_type": "neuronx-cc",
|
26 |
+
"compiler_version": "2.17.194.0+d312836f",
|
27 |
+
"num_cores": 2,
|
28 |
+
"sequence_length": 4096,
|
29 |
+
"task": "text-generation"
|
30 |
+
},
|
31 |
+
"num_attention_heads": 32,
|
32 |
+
"num_hidden_layers": 40,
|
33 |
+
"num_key_value_heads": 8,
|
34 |
+
"residual_multiplier": 0.22,
|
35 |
+
"rms_norm_eps": 1e-05,
|
36 |
+
"rope_scaling": null,
|
37 |
+
"rope_theta": 5000000.0,
|
38 |
+
"tie_word_embeddings": true,
|
39 |
+
"use_cache": true,
|
40 |
+
"vocab_size": 49155
|
41 |
+
}
|
neuronxcc-2.17.194.0+d312836f/MODULE_5b7940618fd1ad4a8869+613edded/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
["--target=trn1", "--model-type=transformer", "--auto-cast=none", "--execute-repetition=1"]
|
neuronxcc-2.17.194.0+d312836f/MODULE_5b7940618fd1ad4a8869+613edded/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_5b7940618fd1ad4a8869+613edded/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dbbd2dee034a69d9882a7124d7c3bd00c745d60ee5f2c2e93e6fdedb9d1c4d2e
|
3 |
+
size 595152
|
neuronxcc-2.17.194.0+d312836f/MODULE_5b7940618fd1ad4a8869+613edded/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:205367ddd085030477ad07a6f2ca8a1d8978e8af2709ee80c3e75ff2d5820170
|
3 |
+
size 3851264
|
neuronxcc-2.17.194.0+d312836f/MODULE_9f5a907ba7db9a6a1e51+613edded/compile_flags.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
["--target=trn1", "--model-type=transformer", "--auto-cast=none", "--execute-repetition=1"]
|
neuronxcc-2.17.194.0+d312836f/MODULE_9f5a907ba7db9a6a1e51+613edded/model.done
ADDED
File without changes
|
neuronxcc-2.17.194.0+d312836f/MODULE_9f5a907ba7db9a6a1e51+613edded/model.hlo_module.pb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:79ec2082448fa44639e1b1600690f18de8d308d4f5aed9617c73d9628277700b
|
3 |
+
size 466083
|
neuronxcc-2.17.194.0+d312836f/MODULE_9f5a907ba7db9a6a1e51+613edded/model.neff
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6e521eb20fc3ac714468b88b888069e72798b8baba0d7e7a80a77f259c91cc68
|
3 |
+
size 7076864
|