Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
optimum-internal-testing-user commited on
Commit
e637a67
·
verified ·
1 Parent(s): 9b70298

Synchronizing local compiler cache.

Browse files
Files changed (6) hide show
  1. neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/gemma3_text/unsloth/gemma-3-270m-it/ba3dcbfc30422641452b.json +90 -0
  2. neuronxcc-2.26.6360.0+6f180f47/MODULE_35250cfcf871d7aba74d+84337dd9/model.hlo_module.pb +1 -1
  3. neuronxcc-2.26.6360.0+6f180f47/MODULE_35250cfcf871d7aba74d+84337dd9/model.neff +2 -2
  4. neuronxcc-2.26.6360.0+6f180f47/MODULE_35250cfcf871d7aba74d+84337dd9/wrapped_neff.hlo +2 -2
  5. neuronxcc-2.26.6360.0+6f180f47/MODULE_8b646f91e023995ffc39+519f203d/model.hlo_module.pb +1 -1
  6. neuronxcc-2.26.6360.0+6f180f47/MODULE_8b646f91e023995ffc39+519f203d/model.neff +2 -2
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/gemma3_text/unsloth/gemma-3-270m-it/ba3dcbfc30422641452b.json ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 1024,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 1024,
56
+ "neuronxcc_version": "2.26.6360.0+6f180f47",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.7.dev0",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "prefill_chunk_size": 0,
62
+ "sequence_length": 1024,
63
+ "speculation_length": 0,
64
+ "start_rank_id": 0,
65
+ "target": "trn1",
66
+ "torch_dtype": "bfloat16",
67
+ "tp_degree": 2
68
+ },
69
+ "num_attention_heads": 4,
70
+ "num_hidden_layers": 18,
71
+ "num_key_value_heads": 1,
72
+ "query_pre_attn_scalar": 256,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_parameters": {
75
+ "full_attention": {
76
+ "rope_theta": 1000000.0,
77
+ "rope_type": "default"
78
+ },
79
+ "sliding_attention": {
80
+ "rope_theta": 10000.0,
81
+ "rope_type": "default"
82
+ }
83
+ },
84
+ "sliding_window": 512,
85
+ "tie_word_embeddings": true,
86
+ "unsloth_fixed": true,
87
+ "use_bidirectional_attention": false,
88
+ "use_cache": true,
89
+ "vocab_size": 262144
90
+ }
neuronxcc-2.26.6360.0+6f180f47/MODULE_35250cfcf871d7aba74d+84337dd9/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7f31b08b9539a8f8ac304cde0638c1c1282bde95deadf397410ca7caead0e8cf
3
  size 546797
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55edff43f740db35902d959551e9f2bb79fdd38a5c47c5b77cc512871a0df8c6
3
  size 546797
neuronxcc-2.26.6360.0+6f180f47/MODULE_35250cfcf871d7aba74d+84337dd9/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cd9a01f937d665117afc61b72f54a920765ac67412202b2cf9eff12647fd5aa9
3
- size 1076224
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea3cb2c5fe0efaa15a6dab00175e53b585260e083641c7c093019b0005298d70
3
+ size 1065984
neuronxcc-2.26.6360.0+6f180f47/MODULE_35250cfcf871d7aba74d+84337dd9/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3a877c4b4a2a7324964fb8f8f19a61083f268f03409652af5fe28763fdbc7a59
3
- size 1198613
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:672be91200b42617e448f538405b332b28beb9daee5bf8cfe785d1c60f3ad417
3
+ size 1188373
neuronxcc-2.26.6360.0+6f180f47/MODULE_8b646f91e023995ffc39+519f203d/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:91c3b9884f8b6e92d76c34dedaf26baefbea564817b637040e2a712ee1475b8a
3
  size 556498
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a66dd17f88388280539ed730ab4508a3a9b88e899dd2cfe5544375227abba5c4
3
  size 556498
neuronxcc-2.26.6360.0+6f180f47/MODULE_8b646f91e023995ffc39+519f203d/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0be9b4adb5d4c0bcd0d1bdd13473874802491c899b1a76be391ed4034a2e167d
3
- size 1618944
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e5bbdb572eb38380d93dbe467d578d4018471e3e0169260b183479192213e64
3
+ size 1608704