dacorvo HF Staff commited on
Commit
d0a4c23
·
verified ·
1 Parent(s): 49bd663

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -15390,3 +15390,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_fdf4978bc53e854ddaa5+a32116a7/model.neff
15390
  neuronxcc-2.21.33363.0+82129205/MODULE_040c216abb7ee9c1e67e+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
15391
  neuronxcc-2.21.33363.0+82129205/MODULE_040c216abb7ee9c1e67e+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
15392
  neuronxcc-2.21.33363.0+82129205/MODULE_d6685031decff9f46567+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
15390
  neuronxcc-2.21.33363.0+82129205/MODULE_040c216abb7ee9c1e67e+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
15391
  neuronxcc-2.21.33363.0+82129205/MODULE_040c216abb7ee9c1e67e+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
15392
  neuronxcc-2.21.33363.0+82129205/MODULE_d6685031decff9f46567+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
15393
+ neuronxcc-2.21.33363.0+82129205/MODULE_7019a22cb8d3eb726aa4+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
15394
+ neuronxcc-2.21.33363.0+82129205/MODULE_7019a22cb8d3eb726aa4+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
15395
+ neuronxcc-2.21.33363.0+82129205/MODULE_bc53d3c124aa43498cbe+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/llama/meta-llama/Llama-2-7b-hf/7ce6a49194e695641d58.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "meta-llama/Llama-2-7b-hf",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "float16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 11008,
16
+ "max_position_embeddings": 4096,
17
+ "mlp_bias": false,
18
+ "model_type": "llama",
19
+ "neuron": {
20
+ "_serialized_key": "NxDNeuronConfig",
21
+ "batch_size": 1,
22
+ "capacity_factor": null,
23
+ "checkpoint_id": "meta-llama/Llama-2-7b-hf",
24
+ "checkpoint_revision": "01c7f73d771dfac7d292323805ebc428287df4f9",
25
+ "continuous_batching": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "max_batch_size": 1,
31
+ "max_context_length": 2048,
32
+ "max_topk": 256,
33
+ "n_active_tokens": 2048,
34
+ "neuronxcc_version": "2.21.33363.0+82129205",
35
+ "on_device_sampling": true,
36
+ "optimum_neuron_version": "0.4.4.dev2",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "sequence_length": 2048,
40
+ "speculation_length": 0,
41
+ "start_rank_id": 0,
42
+ "target": "trn1",
43
+ "torch_dtype": "float16",
44
+ "tp_degree": 2
45
+ },
46
+ "num_attention_heads": 32,
47
+ "num_hidden_layers": 32,
48
+ "num_key_value_heads": 32,
49
+ "pretraining_tp": 1,
50
+ "rms_norm_eps": 1e-05,
51
+ "rope_scaling": null,
52
+ "rope_theta": 10000.0,
53
+ "tie_word_embeddings": false,
54
+ "use_cache": true,
55
+ "vocab_size": 32000
56
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_7019a22cb8d3eb726aa4+a02c3a36/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.33363.0+82129205/MODULE_7019a22cb8d3eb726aa4+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_7019a22cb8d3eb726aa4+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a46d6bf244a21be35ebec272cc402759b968c2fcc08e848a689143013af4298
3
+ size 698771
neuronxcc-2.21.33363.0+82129205/MODULE_7019a22cb8d3eb726aa4+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75e89b49b0342d4ad86da6a44ed064beee9a4742e1d9600f715f201ab73e0c85
3
+ size 10834944
neuronxcc-2.21.33363.0+82129205/MODULE_7019a22cb8d3eb726aa4+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f560e04da95924839e196e2876ada3e77d6016390cfc6e449c33117f91a13dd9
3
+ size 10982639
neuronxcc-2.21.33363.0+82129205/MODULE_bc53d3c124aa43498cbe+24129607/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/context_encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_bc53d3c124aa43498cbe+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_bc53d3c124aa43498cbe+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76f49dac94da97949dc3e5e2fc9d2271494f91a8ddf48715b97f17ebef911865
3
+ size 653427
neuronxcc-2.21.33363.0+82129205/MODULE_bc53d3c124aa43498cbe+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3f79ac4e46264c955548448f876d793575ca8d280462d697e9cdbb96e158810
3
+ size 2857984