Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
709
dacorvo HF Staff commited on
Commit
2a636d2
·
verified ·
1 Parent(s): 10badbe

Synchronizing local compiler cache.

Browse files
Files changed (12) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/dc66a5cddd5aadfa2f284f1dbe3bf39f00874e7eecc249e05bd69f8311cfff65/ccfef6d54e04dd97eda3.json +57 -0
  3. neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/llama/meta-llama/Llama-2-7b-hf/ccfef6d54e04dd97eda3.json +57 -0
  4. neuronxcc-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/compile_flags.json +1 -0
  5. neuronxcc-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/model.done +0 -0
  6. neuronxcc-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/model.hlo_module.pb +3 -0
  7. neuronxcc-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/model.neff +3 -0
  8. neuronxcc-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/wrapped_neff.hlo +3 -0
  9. neuronxcc-2.26.6360.0+6f180f47/MODULE_f5812595797c01ed0ee2+12dd9119/compile_flags.json +1 -0
  10. neuronxcc-2.26.6360.0+6f180f47/MODULE_f5812595797c01ed0ee2+12dd9119/model.done +0 -0
  11. neuronxcc-2.26.6360.0+6f180f47/MODULE_f5812595797c01ed0ee2+12dd9119/model.hlo_module.pb +3 -0
  12. neuronxcc-2.26.6360.0+6f180f47/MODULE_f5812595797c01ed0ee2+12dd9119/model.neff +3 -0
.gitattributes CHANGED
@@ -17882,3 +17882,6 @@ neuronxcc-2.26.6360.0+6f180f47/MODULE_9f6d76d2ac9f4ffeb17a+84337dd9/wrapped_neff
17882
  neuronxcc-2.26.6360.0+6f180f47/MODULE_af9fa19a0af67e193d38+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
17883
  neuronxcc-2.26.6360.0+6f180f47/MODULE_bf02e6e8d85e62734a52+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17884
  neuronxcc-2.26.6360.0+6f180f47/MODULE_bf02e6e8d85e62734a52+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
17882
  neuronxcc-2.26.6360.0+6f180f47/MODULE_af9fa19a0af67e193d38+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
17883
  neuronxcc-2.26.6360.0+6f180f47/MODULE_bf02e6e8d85e62734a52+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17884
  neuronxcc-2.26.6360.0+6f180f47/MODULE_bf02e6e8d85e62734a52+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
17885
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17886
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
17887
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_f5812595797c01ed0ee2+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/dc66a5cddd5aadfa2f284f1dbe3bf39f00874e7eecc249e05bd69f8311cfff65/ccfef6d54e04dd97eda3.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "2.26.6360.0+6f180f47",
35
+ "on_device_sampling": true,
36
+ "optimum_neuron_version": "0.4.7.dev0",
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": "float16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 32,
48
+ "num_hidden_layers": 32,
49
+ "num_key_value_heads": 32,
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-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/llama/meta-llama/Llama-2-7b-hf/ccfef6d54e04dd97eda3.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "2.26.6360.0+6f180f47",
35
+ "on_device_sampling": true,
36
+ "optimum_neuron_version": "0.4.7.dev0",
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": "float16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 32,
48
+ "num_hidden_layers": 32,
49
+ "num_key_value_heads": 32,
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-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/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", "--verbose=35", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc38cfa1455e047d662946b4a59344007e4d5becd89b206525f08e88d39ad278
3
+ size 3030636
neuronxcc-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:295a2090b31d65cb3fe491c478b3aff261ffd8e8f9c8f247d9ef557ad5092331
3
+ size 8899584
neuronxcc-2.26.6360.0+6f180f47/MODULE_a6cbba3c82e366a2748c+84337dd9/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5063ca54023ccc2bc5fda2366dfb06fe4545f058e22df42f57e4a39ea295fba
3
+ size 9047055
neuronxcc-2.26.6360.0+6f180f47/MODULE_f5812595797c01ed0ee2+12dd9119/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", "--verbose=35", "--logfile=/tmp/nxd_model/chunked_prefill/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.26.6360.0+6f180f47/MODULE_f5812595797c01ed0ee2+12dd9119/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_f5812595797c01ed0ee2+12dd9119/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8b6f5f89482aa08cf716a14ba167f629c0684a3913684d25df784ffe6da336e
3
+ size 4082828
neuronxcc-2.26.6360.0+6f180f47/MODULE_f5812595797c01ed0ee2+12dd9119/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:113d49753510425435a09e593d561d51661afc9ff47dd5ab06b28689a3c8a8ef
3
+ size 3544064