Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
dacorvo HF Staff commited on
Commit
0225a9e
·
verified ·
1 Parent(s): e74ec7b

Synchronizing local compiler cache.

Browse files
Files changed (12) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.23.6484.0+3b612583/0_REGISTRY/0.4.6.dev3/4cb7aff9e2a15c151396f2b684013e39d6739f0dec83e5c9dabbfe9d5fcf77b7/4d3dbc5d6f561b83ae44.json +83 -0
  3. neuronxcc-2.23.6484.0+3b612583/0_REGISTRY/0.4.6.dev3/qwen2/Qwen/Qwen2.5-0.5B/4d3dbc5d6f561b83ae44.json +83 -0
  4. neuronxcc-2.23.6484.0+3b612583/MODULE_38e62965360cbacf1bd8+519f203d/compile_flags.json +1 -0
  5. neuronxcc-2.23.6484.0+3b612583/MODULE_38e62965360cbacf1bd8+519f203d/model.done +0 -0
  6. neuronxcc-2.23.6484.0+3b612583/MODULE_38e62965360cbacf1bd8+519f203d/model.hlo_module.pb +3 -0
  7. neuronxcc-2.23.6484.0+3b612583/MODULE_38e62965360cbacf1bd8+519f203d/model.neff +3 -0
  8. neuronxcc-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+84337dd9/compile_flags.json +1 -0
  9. neuronxcc-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+84337dd9/model.done +0 -0
  10. neuronxcc-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+84337dd9/model.hlo_module.pb +3 -0
  11. neuronxcc-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+84337dd9/model.neff +3 -0
  12. neuronxcc-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+84337dd9/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -7304,3 +7304,6 @@ neuronxcc-2.23.6484.0+3b612583/MODULE_b0287a0a8f37672e6b20+faf01bbd/wrapped_neff
7304
  neuronxcc-2.23.6484.0+3b612583/MODULE_2d57017f9e625202a13c+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
7305
  neuronxcc-2.23.6484.0+3b612583/MODULE_fa3c837be0618d501d18+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
7306
  neuronxcc-2.23.6484.0+3b612583/MODULE_fa3c837be0618d501d18+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
7304
  neuronxcc-2.23.6484.0+3b612583/MODULE_2d57017f9e625202a13c+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
7305
  neuronxcc-2.23.6484.0+3b612583/MODULE_fa3c837be0618d501d18+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
7306
  neuronxcc-2.23.6484.0+3b612583/MODULE_fa3c837be0618d501d18+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7307
+ neuronxcc-2.23.6484.0+3b612583/MODULE_38e62965360cbacf1bd8+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
7308
+ neuronxcc-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
7309
+ neuronxcc-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.23.6484.0+3b612583/0_REGISTRY/0.4.6.dev3/4cb7aff9e2a15c151396f2b684013e39d6739f0dec83e5c9dabbfe9d5fcf77b7/4d3dbc5d6f561b83ae44.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen2.5-0.5B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "dtype": "bfloat16",
10
+ "hidden_act": "silu",
11
+ "hidden_size": 896,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 4864,
14
+ "layer_types": [
15
+ "full_attention",
16
+ "full_attention",
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
+ ],
40
+ "max_position_embeddings": 32768,
41
+ "max_window_layers": 24,
42
+ "model_type": "qwen2",
43
+ "neuron": {
44
+ "_serialized_key": "NxDNeuronConfig",
45
+ "batch_size": 4,
46
+ "capacity_factor": null,
47
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
48
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
49
+ "continuous_batching": true,
50
+ "ep_degree": 1,
51
+ "fused_qkv": false,
52
+ "glu_mlp": true,
53
+ "local_ranks_size": 2,
54
+ "max_batch_size": 4,
55
+ "max_context_length": 1024,
56
+ "max_topk": 256,
57
+ "n_active_tokens": 1024,
58
+ "neuronxcc_version": "2.23.6484.0+3b612583",
59
+ "on_device_sampling": false,
60
+ "optimum_neuron_version": "0.4.6.dev3",
61
+ "output_logits": false,
62
+ "pp_degree": 1,
63
+ "prefill_chunk_size": 0,
64
+ "sequence_length": 1024,
65
+ "speculation_length": 0,
66
+ "start_rank_id": 0,
67
+ "target": "trn1",
68
+ "torch_dtype": "bfloat16",
69
+ "tp_degree": 2
70
+ },
71
+ "num_attention_heads": 14,
72
+ "num_hidden_layers": 24,
73
+ "num_key_value_heads": 2,
74
+ "rms_norm_eps": 1e-06,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": null,
78
+ "tie_word_embeddings": true,
79
+ "use_cache": true,
80
+ "use_mrope": false,
81
+ "use_sliding_window": false,
82
+ "vocab_size": 151936
83
+ }
neuronxcc-2.23.6484.0+3b612583/0_REGISTRY/0.4.6.dev3/qwen2/Qwen/Qwen2.5-0.5B/4d3dbc5d6f561b83ae44.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen2.5-0.5B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "dtype": "bfloat16",
10
+ "hidden_act": "silu",
11
+ "hidden_size": 896,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 4864,
14
+ "layer_types": [
15
+ "full_attention",
16
+ "full_attention",
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
+ ],
40
+ "max_position_embeddings": 32768,
41
+ "max_window_layers": 24,
42
+ "model_type": "qwen2",
43
+ "neuron": {
44
+ "_serialized_key": "NxDNeuronConfig",
45
+ "batch_size": 4,
46
+ "capacity_factor": null,
47
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
48
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
49
+ "continuous_batching": true,
50
+ "ep_degree": 1,
51
+ "fused_qkv": false,
52
+ "glu_mlp": true,
53
+ "local_ranks_size": 2,
54
+ "max_batch_size": 4,
55
+ "max_context_length": 1024,
56
+ "max_topk": 256,
57
+ "n_active_tokens": 1024,
58
+ "neuronxcc_version": "2.23.6484.0+3b612583",
59
+ "on_device_sampling": false,
60
+ "optimum_neuron_version": "0.4.6.dev3",
61
+ "output_logits": false,
62
+ "pp_degree": 1,
63
+ "prefill_chunk_size": 0,
64
+ "sequence_length": 1024,
65
+ "speculation_length": 0,
66
+ "start_rank_id": 0,
67
+ "target": "trn1",
68
+ "torch_dtype": "bfloat16",
69
+ "tp_degree": 2
70
+ },
71
+ "num_attention_heads": 14,
72
+ "num_hidden_layers": 24,
73
+ "num_key_value_heads": 2,
74
+ "rms_norm_eps": 1e-06,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": null,
78
+ "tie_word_embeddings": true,
79
+ "use_cache": true,
80
+ "use_mrope": false,
81
+ "use_sliding_window": false,
82
+ "vocab_size": 151936
83
+ }
neuronxcc-2.23.6484.0+3b612583/MODULE_38e62965360cbacf1bd8+519f203d/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/context_encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.23.6484.0+3b612583/MODULE_38e62965360cbacf1bd8+519f203d/model.done ADDED
File without changes
neuronxcc-2.23.6484.0+3b612583/MODULE_38e62965360cbacf1bd8+519f203d/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6e93bb15ba173c27fa18a4bd7f6a6941e8433df7145d0595d67c59ae541a0ed
3
+ size 1977549
neuronxcc-2.23.6484.0+3b612583/MODULE_38e62965360cbacf1bd8+519f203d/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5ef72a9119bba1dff1c4330faac76a2f17fab689b6f4de72e9f416ec003f418
3
+ size 4035584
neuronxcc-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+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-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+84337dd9/model.done ADDED
File without changes
neuronxcc-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+84337dd9/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47bcbde751285ab0c13487ab48e80bf27cf731a23ec9dac4ab90c5dee36393ad
3
+ size 2103950
neuronxcc-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+84337dd9/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:109c69a0beaa38677969e21ce757f9213c96165670e317cdcd04cea305a995ac
3
+ size 1977344
neuronxcc-2.23.6484.0+3b612583/MODULE_5092f2052a543a2f5d04+84337dd9/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28b5882ffc417356be2354a8f673cf46bd84b8d2aed40fef3b571f554417068e
3
+ size 2133535