Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
611
dacorvo HF Staff commited on
Commit
009561d
·
verified ·
1 Parent(s): c7914f2

Synchronizing local compiler cache.

Browse files
Files changed (29) hide show
  1. .gitattributes +8 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/compile_flags.json +1 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/model.done +0 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/model.hlo_module.pb +3 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/model.neff +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/wrapped_neff.hlo +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_4cc7fbab150b94c47fe7+be13b572/compile_flags.json +1 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_4cc7fbab150b94c47fe7+be13b572/model.hlo_module.pb +3 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_4cc7fbab150b94c47fe7+be13b572/model.log +3 -0
  10. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.3.dev2/qwen3/Qwen/Qwen3-1.7B/6405bf28ede725e12a89.json +87 -0
  11. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.3.dev2/qwen3/Qwen/Qwen3-1.7B/7857eae3bcd2e9e1f40b.json +87 -0
  12. neuronxcc-2.21.33363.0+82129205/MODULE_04b8a50f3449a2352a25+a02c3a36/compile_flags.json +1 -0
  13. neuronxcc-2.21.33363.0+82129205/MODULE_04b8a50f3449a2352a25+a02c3a36/model.done +0 -0
  14. neuronxcc-2.21.33363.0+82129205/MODULE_04b8a50f3449a2352a25+a02c3a36/model.hlo_module.pb +3 -0
  15. neuronxcc-2.21.33363.0+82129205/MODULE_04b8a50f3449a2352a25+a02c3a36/model.neff +3 -0
  16. neuronxcc-2.21.33363.0+82129205/MODULE_04b8a50f3449a2352a25+a02c3a36/wrapped_neff.hlo +3 -0
  17. neuronxcc-2.21.33363.0+82129205/MODULE_224bcebb9f0386b8dc10+24129607/compile_flags.json +1 -0
  18. neuronxcc-2.21.33363.0+82129205/MODULE_224bcebb9f0386b8dc10+24129607/model.done +0 -0
  19. neuronxcc-2.21.33363.0+82129205/MODULE_224bcebb9f0386b8dc10+24129607/model.hlo_module.pb +3 -0
  20. neuronxcc-2.21.33363.0+82129205/MODULE_224bcebb9f0386b8dc10+24129607/model.neff +3 -0
  21. neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/compile_flags.json +1 -0
  22. neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/model.done +0 -0
  23. neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/model.hlo_module.pb +3 -0
  24. neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/model.neff +3 -0
  25. neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/wrapped_neff.hlo +3 -0
  26. neuronxcc-2.21.33363.0+82129205/MODULE_ce103554453087bfd021+a32116a7/compile_flags.json +1 -0
  27. neuronxcc-2.21.33363.0+82129205/MODULE_ce103554453087bfd021+a32116a7/model.done +0 -0
  28. neuronxcc-2.21.33363.0+82129205/MODULE_ce103554453087bfd021+a32116a7/model.hlo_module.pb +3 -0
  29. neuronxcc-2.21.33363.0+82129205/MODULE_ce103554453087bfd021+a32116a7/model.neff +3 -0
.gitattributes CHANGED
@@ -13831,3 +13831,11 @@ neuronxcc-2.21.33363.0+82129205/MODULE_1280d8f2f3240abd5752+a02c3a36/wrapped_nef
13831
  neuronxcc-2.21.33363.0+82129205/MODULE_0ca0b7d2c1246aa065a0+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
13832
  neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
13833
  neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
13831
  neuronxcc-2.21.33363.0+82129205/MODULE_0ca0b7d2c1246aa065a0+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
13832
  neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
13833
  neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
13834
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/model.neff filter=lfs diff=lfs merge=lfs -text
13835
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
13836
+ neuronxcc-2.21.33363.0+82129205/MODULE_04b8a50f3449a2352a25+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
13837
+ neuronxcc-2.21.33363.0+82129205/MODULE_04b8a50f3449a2352a25+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
13838
+ neuronxcc-2.21.33363.0+82129205/MODULE_224bcebb9f0386b8dc10+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
13839
+ neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/model.neff filter=lfs diff=lfs merge=lfs -text
13840
+ neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
13841
+ neuronxcc-2.21.33363.0+82129205/MODULE_ce103554453087bfd021+a32116a7/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e1028849e83cf491d3dba614c6f80394150d911a91d5aea12afb950177f59f0
3
+ size 2072225
neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1de726a89644d3d786eb57dfee935c105e879e841f6fab5d2662ae220f52526
3
+ size 1045504
neuronxcc-2.21.18209.0+043b1bf7/MODULE_082293dad00be4d3c3fc+0b01cb42/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2589740fc5b9de185c880155d4e1de915c634da0bbd7ad1cb3f95eafa21ed77a
3
+ size 1320326
neuronxcc-2.21.18209.0+043b1bf7/MODULE_4cc7fbab150b94c47fe7+be13b572/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_4cc7fbab150b94c47fe7+be13b572/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:766c7a7ffb36f919502eedbcc5129962f836a8f6520fbd82858a7b9df483fa3b
3
+ size 34606263
neuronxcc-2.21.18209.0+043b1bf7/MODULE_4cc7fbab150b94c47fe7+be13b572/model.log ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ Failed compilation with ['neuronx-cc', 'compile', '--framework=XLA', '/tmp/nxd_model/context_encoding_model/_tp0_bk0/model.MODULE_4cc7fbab150b94c47fe7+be13b572.hlo_module.pb', '--output', '/tmp/nxd_model/context_encoding_model/_tp0_bk0/model.MODULE_4cc7fbab150b94c47fe7+be13b572.neff', '--target=trn2', '--enable-saturate-infinity', '--enable-mixed-precision-accumulation', '--model-type', 'transformer', '-O1', '--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2', '--auto-cast=none', '--internal-enable-dge-levels', 'vector_dynamic_offsets', '--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt', '--verbose=35']: [XCG815] Estimated peak HBM usage (27.758GB) exceeds 24GB. Neff might be unable to load on chip. If you believe this estimation to be inaccurate, you can disable the check using: `--internal-backend-options=' --disable-hbm-usage-check '` - Please open a support ticket at https://github.com/aws-neuron/aws-neuron-sdk/issues/new. You may also be able to obtain more information using the 'XLA_IR_DEBUG' and 'XLA_HLO_DEBUG' environment variables.
2
+ 2025-10-21T16:24:32Z Non-signal exit. Backend exited with code 1 and stderr: [XCG815] Estimated peak HBM usage (27.758GB) exceeds 24GB. Neff might be unable to load on chip. If you believe this estimation to be inaccurate, you can disable the check using: `--internal-backend-options=' --disable-hbm-usage-check '` - Please open a support ticket at https://github.com/aws-neuron/aws-neuron-sdk/issues/new. You may also be able to obtain more information using the 'XLA_IR_DEBUG' and 'XLA_HLO_DEBUG' environment variables.
3
+
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.3.dev2/qwen3/Qwen/Qwen3-1.7B/6405bf28ede725e12a89.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-1.7B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 2048,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 6144,
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
+ ],
46
+ "max_position_embeddings": 40960,
47
+ "max_window_layers": 28,
48
+ "model_type": "qwen3",
49
+ "neuron": {
50
+ "_serialized_key": "NxDNeuronConfig",
51
+ "batch_size": 4,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-1.7B",
54
+ "checkpoint_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
55
+ "continuous_batching": true,
56
+ "ep_degree": 1,
57
+ "fused_qkv": true,
58
+ "glu_mlp": true,
59
+ "local_ranks_size": 2,
60
+ "max_batch_size": 4,
61
+ "max_context_length": 2048,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 2048,
64
+ "neuronxcc_version": "2.21.33363.0+82129205",
65
+ "on_device_sampling": false,
66
+ "optimum_neuron_version": "0.4.3.dev2",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 2048,
70
+ "speculation_length": 0,
71
+ "start_rank_id": 0,
72
+ "target": "trn2",
73
+ "torch_dtype": "bfloat16",
74
+ "tp_degree": 2
75
+ },
76
+ "num_attention_heads": 16,
77
+ "num_hidden_layers": 28,
78
+ "num_key_value_heads": 8,
79
+ "rms_norm_eps": 1e-06,
80
+ "rope_scaling": null,
81
+ "rope_theta": 1000000,
82
+ "sliding_window": null,
83
+ "tie_word_embeddings": true,
84
+ "use_cache": true,
85
+ "use_sliding_window": false,
86
+ "vocab_size": 151936
87
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.3.dev2/qwen3/Qwen/Qwen3-1.7B/7857eae3bcd2e9e1f40b.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-1.7B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 2048,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 6144,
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
+ ],
46
+ "max_position_embeddings": 40960,
47
+ "max_window_layers": 28,
48
+ "model_type": "qwen3",
49
+ "neuron": {
50
+ "_serialized_key": "NxDNeuronConfig",
51
+ "batch_size": 4,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-1.7B",
54
+ "checkpoint_revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e",
55
+ "continuous_batching": true,
56
+ "ep_degree": 1,
57
+ "fused_qkv": true,
58
+ "glu_mlp": true,
59
+ "local_ranks_size": 2,
60
+ "max_batch_size": 4,
61
+ "max_context_length": 2048,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 2048,
64
+ "neuronxcc_version": "2.21.33363.0+82129205",
65
+ "on_device_sampling": false,
66
+ "optimum_neuron_version": "0.4.3.dev2",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 2048,
70
+ "speculation_length": 0,
71
+ "start_rank_id": 0,
72
+ "target": "trn1",
73
+ "torch_dtype": "bfloat16",
74
+ "tp_degree": 2
75
+ },
76
+ "num_attention_heads": 16,
77
+ "num_hidden_layers": 28,
78
+ "num_key_value_heads": 8,
79
+ "rms_norm_eps": 1e-06,
80
+ "rope_scaling": null,
81
+ "rope_theta": 1000000,
82
+ "sliding_window": null,
83
+ "tie_word_embeddings": true,
84
+ "use_cache": true,
85
+ "use_sliding_window": false,
86
+ "vocab_size": 151936
87
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_04b8a50f3449a2352a25+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_04b8a50f3449a2352a25+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_04b8a50f3449a2352a25+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b73e1e924bcc1ed0c2d1355b0fd97e5e8e9f5327a69d4d90cb7839c18fe367a6
3
+ size 702870
neuronxcc-2.21.33363.0+82129205/MODULE_04b8a50f3449a2352a25+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec330152e867baa711d7b5d4efb5daa09fe7f1cadae4d080f860749b7a59cbb9
3
+ size 3912704
neuronxcc-2.21.33363.0+82129205/MODULE_04b8a50f3449a2352a25+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dcbe10f71b9e88b08abf39ce3357068772e5344c68fa6e1bd5c50c6fbdbd8a9
3
+ size 4049704
neuronxcc-2.21.33363.0+82129205/MODULE_224bcebb9f0386b8dc10+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_224bcebb9f0386b8dc10+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_224bcebb9f0386b8dc10+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77b9348784becba5059375c0eb95233dcbdf93043535fa50bb53d57513feb444
3
+ size 693336
neuronxcc-2.21.33363.0+82129205/MODULE_224bcebb9f0386b8dc10+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01b50f52d481e180ab8086383b5ce2715b9b4f117f1472f9b58d4514b9dac4a8
3
+ size 17705984
neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/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=2", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11a6c3cc153eee63b1e46c20941f0830756ecd30f9bb8f1cbac218b6244b17c7
3
+ size 702870
neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d5c0d9e2e6d1d3770c6ba7eb61ea6b08ea58c5997250122c1363af1b05bff61
3
+ size 4731904
neuronxcc-2.21.33363.0+82129205/MODULE_94ebaa8c44093dcf8268+ac10809c/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e887beae4b1a6939a48242f2d0ac0f433589d80abb5e347d6546a5e565e078a
3
+ size 4869884
neuronxcc-2.21.33363.0+82129205/MODULE_ce103554453087bfd021+a32116a7/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=2", "--logfile=/tmp/nxd_model/context_encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_ce103554453087bfd021+a32116a7/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_ce103554453087bfd021+a32116a7/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ba4b216de474e3ffe63c5e067cdb6832ad726828112a224097a55bf8052114c
3
+ size 972420
neuronxcc-2.21.33363.0+82129205/MODULE_ce103554453087bfd021+a32116a7/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23f043d57c5ea5d4fa7666cb2b6bdd96a180ca68e6aebcc91923b574d228d102
3
+ size 17572864