Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
625
jburtoft commited on
Commit
5531e20
·
verified ·
1 Parent(s): 5f8bb2a

Synchronizing local compiler cache.

Browse files
Files changed (15) hide show
  1. .gitattributes +4 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.3.0/qwen3/karanps/ChessLM_Qwen3/52efb79a5a97452cc47a.json +111 -0
  3. neuronxcc-2.21.33363.0+82129205/MODULE_835ff986544e4125ef22+ae6a382b/compile_flags.json +1 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_835ff986544e4125ef22+ae6a382b/model.done +0 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_835ff986544e4125ef22+ae6a382b/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_835ff986544e4125ef22+ae6a382b/model.neff +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/compile_flags.json +1 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/model.done +0 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/model.neff +3 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/wrapped_neff.hlo +3 -0
  12. neuronxcc-2.21.33363.0+82129205/MODULE_e0e9037eb566898d9d22+b6452646/compile_flags.json +1 -0
  13. neuronxcc-2.21.33363.0+82129205/MODULE_e0e9037eb566898d9d22+b6452646/model.done +0 -0
  14. neuronxcc-2.21.33363.0+82129205/MODULE_e0e9037eb566898d9d22+b6452646/model.hlo_module.pb +3 -0
  15. neuronxcc-2.21.33363.0+82129205/MODULE_e0e9037eb566898d9d22+b6452646/model.neff +3 -0
.gitattributes CHANGED
@@ -13113,3 +13113,7 @@ neuronxcc-2.21.33363.0+82129205/MODULE_313b6e98dbc76fc074be+690e2d39/model.neff
13113
  neuronxcc-2.21.33363.0+82129205/MODULE_313b6e98dbc76fc074be+690e2d39/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
13114
  neuronxcc-2.21.33363.0+82129205/MODULE_3d885529496fe38a321d+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
13115
  neuronxcc-2.21.33363.0+82129205/MODULE_d68ad03c2c4006bf7b13+b6452646/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
13113
  neuronxcc-2.21.33363.0+82129205/MODULE_313b6e98dbc76fc074be+690e2d39/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
13114
  neuronxcc-2.21.33363.0+82129205/MODULE_3d885529496fe38a321d+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
13115
  neuronxcc-2.21.33363.0+82129205/MODULE_d68ad03c2c4006bf7b13+b6452646/model.neff filter=lfs diff=lfs merge=lfs -text
13116
+ neuronxcc-2.21.33363.0+82129205/MODULE_835ff986544e4125ef22+ae6a382b/model.neff filter=lfs diff=lfs merge=lfs -text
13117
+ neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/model.neff filter=lfs diff=lfs merge=lfs -text
13118
+ neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
13119
+ neuronxcc-2.21.33363.0+82129205/MODULE_e0e9037eb566898d9d22+b6452646/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.3.0/qwen3/karanps/ChessLM_Qwen3/52efb79a5a97452cc47a.json ADDED
@@ -0,0 +1,111 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "karanps/ChessLM_Qwen3",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "float32",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 12288,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention"
53
+ ],
54
+ "max_position_embeddings": 40960,
55
+ "max_window_layers": 36,
56
+ "model_type": "qwen3",
57
+ "neuron": {
58
+ "_serialized_key": "NxDNeuronConfig",
59
+ "async_mode": false,
60
+ "attn_kernel_enabled": false,
61
+ "batch_size": 4,
62
+ "capacity_factor": null,
63
+ "cc_pipeline_tiling_factor": 2,
64
+ "checkpoint_id": "karanps/ChessLM_Qwen3",
65
+ "checkpoint_revision": "e0d57507d96b2be2dd0dc901ecb231dec2dd6330",
66
+ "continuous_batching": true,
67
+ "enable_bucketing": false,
68
+ "ep_degree": 1,
69
+ "flash_decoding_enabled": false,
70
+ "fused_qkv": true,
71
+ "glu_mlp": true,
72
+ "is_chunked_prefill": false,
73
+ "local_ranks_size": 2,
74
+ "logical_nc_config": 1,
75
+ "max_batch_size": 4,
76
+ "max_context_length": 2048,
77
+ "max_topk": 256,
78
+ "mlp_kernel_enabled": false,
79
+ "mlp_kernel_fuse_residual_add": false,
80
+ "n_active_tokens": 2048,
81
+ "neuronxcc_version": "2.21.33363.0+82129205",
82
+ "num_cores_per_group": 1,
83
+ "on_device_sampling": false,
84
+ "optimum_neuron_version": "0.3.0",
85
+ "output_logits": false,
86
+ "padding_side": "right",
87
+ "pp_degree": 1,
88
+ "qk_layernorm": false,
89
+ "qkv_kernel_enabled": false,
90
+ "rpl_reduce_dtype": "bfloat16",
91
+ "sequence_length": 2048,
92
+ "sequence_parallel_enabled": false,
93
+ "speculation_length": 0,
94
+ "start_rank_id": 0,
95
+ "target": null,
96
+ "torch_dtype": "bfloat16",
97
+ "tp_degree": 2,
98
+ "vocab_parallel": false
99
+ },
100
+ "num_attention_heads": 32,
101
+ "num_hidden_layers": 36,
102
+ "num_key_value_heads": 8,
103
+ "rms_norm_eps": 1e-06,
104
+ "rope_scaling": null,
105
+ "rope_theta": 1000000,
106
+ "sliding_window": null,
107
+ "tie_word_embeddings": false,
108
+ "use_cache": true,
109
+ "use_sliding_window": false,
110
+ "vocab_size": 151936
111
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_835ff986544e4125ef22+ae6a382b/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options='--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --verify-hlo=false' --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt"
neuronxcc-2.21.33363.0+82129205/MODULE_835ff986544e4125ef22+ae6a382b/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_835ff986544e4125ef22+ae6a382b/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b77aa940bcec695e9b9a84fe3abd8d5e4d8169f7039062ff9a1f5feca32f2797
3
+ size 173259
neuronxcc-2.21.33363.0+82129205/MODULE_835ff986544e4125ef22+ae6a382b/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dec53c4a5d9fd71032cb4271d1845a6c5678821d91f2396f927cc8d1667736ac
3
+ size 6513664
neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--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_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a948f2c60b6b43926353f92af33986b0bb0382a9b0151756233036815f9bddf
3
+ size 904619
neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34e05a59e71a253ad29174efa5eea5813cc1ba07af900eef6704da20b006724c
3
+ size 10220544
neuronxcc-2.21.33363.0+82129205/MODULE_c5d57e15f7d917212c36+690e2d39/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:831136d56a8d00b3747d3939b9f30711bf73083afb2ba710798203e92721bf10
3
+ size 10395651
neuronxcc-2.21.33363.0+82129205/MODULE_e0e9037eb566898d9d22+b6452646/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--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_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_e0e9037eb566898d9d22+b6452646/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_e0e9037eb566898d9d22+b6452646/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9cba0c58bd34380b23cbc55998d0eb27f21e1d6d07ee97547de0c6c504ae3e8d
3
+ size 889438
neuronxcc-2.21.33363.0+82129205/MODULE_e0e9037eb566898d9d22+b6452646/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83a11b7d0b863bd5a68c13dbc48593c5f1c9029b846930d57b527f722b0e1730
3
+ size 4281344