Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
709
dacorvo HF Staff commited on
Commit
eaae793
·
verified ·
1 Parent(s): 40e935a

Synchronizing local compiler cache.

Browse files
Files changed (7) hide show
  1. .gitattributes +1 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/719fe7a9c6374369ebb3c8b70e3ca2e20d29e8d4c6f1965131b188cd23d90d00/2135f0cf0a5095ac538b.json +57 -0
  3. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/llama/TinyLlama/TinyLlama-1.1B-Chat-v1.0/2135f0cf0a5095ac538b.json +57 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_feb2132159f92e3e58e5+6170d8e1/compile_flags.json +1 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_feb2132159f92e3e58e5+6170d8e1/model.done +0 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_feb2132159f92e3e58e5+6170d8e1/model.hlo_module.pb +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_feb2132159f92e3e58e5+6170d8e1/model.neff +3 -0
.gitattributes CHANGED
@@ -17368,3 +17368,4 @@ neuronxcc-2.21.33363.0+82129205/MODULE_ff6e0ad0581f507d369c+6170d8e1/model.neff
17368
  neuronxcc-2.21.33363.0+82129205/MODULE_98da9e5da8480819d736+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
17369
  neuronxcc-2.21.33363.0+82129205/MODULE_5948d8ea6d936e9cc55c+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
17370
  neuronxcc-2.21.33363.0+82129205/MODULE_d608257aadfd064df069+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
17368
  neuronxcc-2.21.33363.0+82129205/MODULE_98da9e5da8480819d736+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
17369
  neuronxcc-2.21.33363.0+82129205/MODULE_5948d8ea6d936e9cc55c+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
17370
  neuronxcc-2.21.33363.0+82129205/MODULE_d608257aadfd064df069+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
17371
+ neuronxcc-2.21.33363.0+82129205/MODULE_feb2132159f92e3e58e5+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/719fe7a9c6374369ebb3c8b70e3ca2e20d29e8d4c6f1965131b188cd23d90d00/2135f0cf0a5095ac538b.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
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": 5632,
16
+ "max_position_embeddings": 2048,
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": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
24
+ "checkpoint_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
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": "0.4.6.dev3",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "prefill_chunk_size": 1024,
40
+ "sequence_length": 2048,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": "trn1",
44
+ "torch_dtype": "bfloat16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 32,
48
+ "num_hidden_layers": 22,
49
+ "num_key_value_heads": 4,
50
+ "pretraining_tp": 1,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": null,
53
+ "rope_theta": 10000.0,
54
+ "tie_word_embeddings": false,
55
+ "use_cache": true,
56
+ "vocab_size": 32000
57
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/llama/TinyLlama/TinyLlama-1.1B-Chat-v1.0/2135f0cf0a5095ac538b.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
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": 5632,
16
+ "max_position_embeddings": 2048,
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": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
24
+ "checkpoint_revision": "fe8a4ea1ffedaf415f4da2f062534de366a451e6",
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": "0.4.6.dev3",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "prefill_chunk_size": 1024,
40
+ "sequence_length": 2048,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": "trn1",
44
+ "torch_dtype": "bfloat16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 32,
48
+ "num_hidden_layers": 22,
49
+ "num_key_value_heads": 4,
50
+ "pretraining_tp": 1,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": null,
53
+ "rope_theta": 10000.0,
54
+ "tie_word_embeddings": false,
55
+ "use_cache": true,
56
+ "vocab_size": 32000
57
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_feb2132159f92e3e58e5+6170d8e1/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/chunked_prefill/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_feb2132159f92e3e58e5+6170d8e1/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_feb2132159f92e3e58e5+6170d8e1/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a30ec35c107128a63b80cfc57670d3a3e8f4bc0d82b672fde54f558ff19e65b0
3
+ size 749291
neuronxcc-2.21.33363.0+82129205/MODULE_feb2132159f92e3e58e5+6170d8e1/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af1215c108165a66f4f242026df3e5687ab611d9ddb56ffd2aea164a634d0e30
3
+ size 25324544