dacorvo HF Staff commited on
Commit
afb02e2
·
verified ·
1 Parent(s): 1b99c98

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -6038,3 +6038,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_cfd70138eb9722ac2255+a02c3a36/model.neff
6038
  neuronxcc-2.21.33363.0+82129205/MODULE_cfd70138eb9722ac2255+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6039
  neuronxcc-2.21.33363.0+82129205/MODULE_dd366124647aeec64074+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
6040
  neuronxcc-2.21.33363.0+82129205/MODULE_dd366124647aeec64074+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
6038
  neuronxcc-2.21.33363.0+82129205/MODULE_cfd70138eb9722ac2255+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6039
  neuronxcc-2.21.33363.0+82129205/MODULE_dd366124647aeec64074+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
6040
  neuronxcc-2.21.33363.0+82129205/MODULE_dd366124647aeec64074+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6041
+ neuronxcc-2.21.33363.0+82129205/MODULE_37d321becc90cb687039+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
6042
+ neuronxcc-2.21.33363.0+82129205/MODULE_9b8221eb10f6b4eb2f68+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
6043
+ neuronxcc-2.21.33363.0+82129205/MODULE_9b8221eb10f6b4eb2f68+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/llama/unsloth/Llama-3.2-1B-Instruct/93fa413575d5ccc52f58.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "unsloth/Llama-3.2-1B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 64,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 2048,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 8192,
16
+ "max_position_embeddings": 131072,
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": "unsloth/Llama-3.2-1B-Instruct",
24
+ "checkpoint_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
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": 8192,
32
+ "max_topk": 256,
33
+ "n_active_tokens": 8192,
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": 8192,
40
+ "speculation_length": 0,
41
+ "start_rank_id": 0,
42
+ "target": "trn1",
43
+ "torch_dtype": "bfloat16",
44
+ "tp_degree": 2
45
+ },
46
+ "num_attention_heads": 32,
47
+ "num_hidden_layers": 16,
48
+ "num_key_value_heads": 8,
49
+ "pretraining_tp": 1,
50
+ "rms_norm_eps": 1e-05,
51
+ "rope_scaling": {
52
+ "factor": 32.0,
53
+ "high_freq_factor": 4.0,
54
+ "low_freq_factor": 1.0,
55
+ "original_max_position_embeddings": 8192,
56
+ "rope_type": "llama3"
57
+ },
58
+ "rope_theta": 500000.0,
59
+ "tie_word_embeddings": true,
60
+ "unsloth_fixed": true,
61
+ "use_cache": true,
62
+ "vocab_size": 128256
63
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_37d321becc90cb687039+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_37d321becc90cb687039+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_37d321becc90cb687039+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b254a5de285d4202654c3861698c1a55558bbae98505f678539066015a5d803a
3
+ size 474402
neuronxcc-2.21.33363.0+82129205/MODULE_37d321becc90cb687039+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02b5479a35276997f52e0427e90139634e919e308bc9747f3ca820dec15d0382
3
+ size 115057664
neuronxcc-2.21.33363.0+82129205/MODULE_9b8221eb10f6b4eb2f68+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_9b8221eb10f6b4eb2f68+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_9b8221eb10f6b4eb2f68+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2560de39994ba4784626257a4845c5ad7af11cf2ce79e7df7ac1e29ba15f33ec
3
+ size 434264
neuronxcc-2.21.33363.0+82129205/MODULE_9b8221eb10f6b4eb2f68+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c2586694f84bb7527a209850d0dd49d48990faacd1d4e4c72602d5e32573172
3
+ size 2366464
neuronxcc-2.21.33363.0+82129205/MODULE_9b8221eb10f6b4eb2f68+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce92ab94bac6c24d2adf4d57bc2afbacfc4ec63abecfe9b3a06f395d6077f1ed
3
+ size 2440551