Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
dacorvo HF Staff commited on
Commit
32bd799
·
verified ·
1 Parent(s): a665fbd

Synchronizing local compiler cache.

Browse files
Files changed (8) hide show
  1. .gitattributes +2 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/441269935591cad8d370e512c0b93cdd2fce6247c40e5a4866d872ee5338b0de/04dbf3b30744c2b2a648.json +81 -0
  3. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/gemma3_text/unsloth/gemma-3-270m-it/04dbf3b30744c2b2a648.json +81 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_1276e3437f32f0068abf+a02c3a36/compile_flags.json +1 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_1276e3437f32f0068abf+a02c3a36/model.done +0 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_1276e3437f32f0068abf+a02c3a36/model.hlo_module.pb +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_1276e3437f32f0068abf+a02c3a36/model.neff +3 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_1276e3437f32f0068abf+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -6977,3 +6977,5 @@ neuronxcc-2.21.33363.0+82129205/MODULE_b883f81328479f8fb829+a02c3a36/model.neff
6977
  neuronxcc-2.21.33363.0+82129205/MODULE_b883f81328479f8fb829+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6978
  neuronxcc-2.21.33363.0+82129205/MODULE_087174b21325f420c318+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
6979
  neuronxcc-2.21.33363.0+82129205/MODULE_087174b21325f420c318+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
6977
  neuronxcc-2.21.33363.0+82129205/MODULE_b883f81328479f8fb829+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6978
  neuronxcc-2.21.33363.0+82129205/MODULE_087174b21325f420c318+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
6979
  neuronxcc-2.21.33363.0+82129205/MODULE_087174b21325f420c318+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6980
+ neuronxcc-2.21.33363.0+82129205/MODULE_1276e3437f32f0068abf+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
6981
+ neuronxcc-2.21.33363.0+82129205/MODULE_1276e3437f32f0068abf+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/441269935591cad8d370e512c0b93cdd2fce6247c40e5a4866d872ee5338b0de/04dbf3b30744c2b2a648.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "unsloth/gemma-3-270m-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "dtype": "bfloat16",
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 640,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 2048,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention"
38
+ ],
39
+ "max_position_embeddings": 32768,
40
+ "model_type": "gemma3_text",
41
+ "neuron": {
42
+ "_serialized_key": "NxDNeuronConfig",
43
+ "batch_size": 4,
44
+ "capacity_factor": null,
45
+ "checkpoint_id": "unsloth/gemma-3-270m-it",
46
+ "checkpoint_revision": "23cf460f6bb16954176b3ddcc8d4f250501458a9",
47
+ "continuous_batching": true,
48
+ "ep_degree": 1,
49
+ "fused_qkv": true,
50
+ "glu_mlp": true,
51
+ "local_ranks_size": 2,
52
+ "max_batch_size": 4,
53
+ "max_context_length": 4096,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 4096,
56
+ "neuronxcc_version": "2.21.33363.0+82129205",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.6.dev2",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "sequence_length": 4096,
62
+ "speculation_length": 0,
63
+ "start_rank_id": 0,
64
+ "target": "trn1",
65
+ "torch_dtype": "bfloat16",
66
+ "tp_degree": 2
67
+ },
68
+ "num_attention_heads": 4,
69
+ "num_hidden_layers": 18,
70
+ "num_key_value_heads": 1,
71
+ "query_pre_attn_scalar": 256,
72
+ "rms_norm_eps": 1e-06,
73
+ "rope_local_base_freq": 10000.0,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": 512,
77
+ "unsloth_fixed": true,
78
+ "use_bidirectional_attention": false,
79
+ "use_cache": true,
80
+ "vocab_size": 262144
81
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/gemma3_text/unsloth/gemma-3-270m-it/04dbf3b30744c2b2a648.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "unsloth/gemma-3-270m-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "dtype": "bfloat16",
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 640,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 2048,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention"
38
+ ],
39
+ "max_position_embeddings": 32768,
40
+ "model_type": "gemma3_text",
41
+ "neuron": {
42
+ "_serialized_key": "NxDNeuronConfig",
43
+ "batch_size": 4,
44
+ "capacity_factor": null,
45
+ "checkpoint_id": "unsloth/gemma-3-270m-it",
46
+ "checkpoint_revision": "23cf460f6bb16954176b3ddcc8d4f250501458a9",
47
+ "continuous_batching": true,
48
+ "ep_degree": 1,
49
+ "fused_qkv": true,
50
+ "glu_mlp": true,
51
+ "local_ranks_size": 2,
52
+ "max_batch_size": 4,
53
+ "max_context_length": 4096,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 4096,
56
+ "neuronxcc_version": "2.21.33363.0+82129205",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.6.dev2",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "sequence_length": 4096,
62
+ "speculation_length": 0,
63
+ "start_rank_id": 0,
64
+ "target": "trn1",
65
+ "torch_dtype": "bfloat16",
66
+ "tp_degree": 2
67
+ },
68
+ "num_attention_heads": 4,
69
+ "num_hidden_layers": 18,
70
+ "num_key_value_heads": 1,
71
+ "query_pre_attn_scalar": 256,
72
+ "rms_norm_eps": 1e-06,
73
+ "rope_local_base_freq": 10000.0,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": 512,
77
+ "unsloth_fixed": true,
78
+ "use_bidirectional_attention": false,
79
+ "use_cache": true,
80
+ "vocab_size": 262144
81
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_1276e3437f32f0068abf+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_1276e3437f32f0068abf+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_1276e3437f32f0068abf+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47c7f226b92eeef5d21729f262c89be08baabef5cb92e026f6fbbb9fcafa46b1
3
+ size 578997
neuronxcc-2.21.33363.0+82129205/MODULE_1276e3437f32f0068abf+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba9316eee2f095f4268baa22375f316e2b22c3adbbc60978f0d79bb282114f7d
3
+ size 1762304
neuronxcc-2.21.33363.0+82129205/MODULE_1276e3437f32f0068abf+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04d0e49e0e456b77f29b9a085aea8fbb222750871a6a29199d44f9c6dd15b8bd
3
+ size 1884693