Synchronizing local compiler cache.
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +12 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/llama/unsloth/Llama-3.2-1B-Instruct/c4a5702b42ed466de9e7.json +63 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/qwen2/Qwen/Qwen2.5-0.5B/df54a6f9b56ca3bdfc41.json +82 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/smollm3/HuggingFaceTB/SmolLM3-3B/1a86e1cfcaf229143d5c.json +134 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_10217061096959125489+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_10244305442015770634+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_10606948783918825529+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_10645643398657092095+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_10661660426924300837+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_10746122569655005679+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_11086318750207148626+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_11909310600244571805+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_11933818254123612383+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1210392327607194823+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_12487216553200321032+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_13085549342645515693+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_13102010590082783346+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_13475273426270778455+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_13728813963059599796+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_13783872956654054643+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_14097776130612710282+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_14155105943057125178+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_14173682194645188821+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_14274833313744358348+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_14469716438804935215+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_14739672283771660808+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_15442663025941492357+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_15557971880365771457+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_15564425602929126510+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_15589838460896944293+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_15654572849171857535+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_15767383571209512795+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_15888836045088309511+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_15929451261464042997+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_16016505958416521648+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_16053163479112702088+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_16204387852795926216+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1656719109221189948+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_16599571375348449904+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1663757798483801648+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_16738296820980389103+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1696825468766062114+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_17476209562158013765+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_17781596253725927902+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1795740353831178306+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1800832390737682969+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1805481651134498710+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_1805728312057401221+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_18070749384531238695+e30acd3a/model.neff +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_18134517905922687964+e30acd3a/model.neff +0 -0
.gitattributes
CHANGED
|
@@ -15564,3 +15564,15 @@ neuronxcc-2.21.33363.0+82129205/MODULE_beeee808ee2c5817dadb+41bea4e3/model.neff
|
|
| 15564 |
neuronxcc-2.21.33363.0+82129205/MODULE_beeee808ee2c5817dadb+41bea4e3/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15565 |
neuronxcc-2.21.33363.0+82129205/MODULE_c01f2cdb2ffaf3dd482e+0c4991c2/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15566 |
neuronxcc-2.21.33363.0+82129205/MODULE_c01f2cdb2ffaf3dd482e+0c4991c2/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15564 |
neuronxcc-2.21.33363.0+82129205/MODULE_beeee808ee2c5817dadb+41bea4e3/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15565 |
neuronxcc-2.21.33363.0+82129205/MODULE_c01f2cdb2ffaf3dd482e+0c4991c2/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15566 |
neuronxcc-2.21.33363.0+82129205/MODULE_c01f2cdb2ffaf3dd482e+0c4991c2/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15567 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_1dab43670aa461ded2e0+3d34d2fe/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15568 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_1dab43670aa461ded2e0+3d34d2fe/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15569 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_229d6c9515064bead177+847f14c9/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15570 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_229d6c9515064bead177+847f14c9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15571 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_2487f697828dacb3ae17+d7f19d7c/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15572 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_2487f697828dacb3ae17+d7f19d7c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15573 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_4f1a7a6fa2f9e9ca209e+9ad87327/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15574 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_4f1a7a6fa2f9e9ca209e+9ad87327/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15575 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_6679a28c4e41ad2ee46a+67189fe6/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15576 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_6679a28c4e41ad2ee46a+67189fe6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15577 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_8edd215cebed1af675cc+6a9d8a9a/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15578 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_8edd215cebed1af675cc+6a9d8a9a/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/llama/unsloth/Llama-3.2-1B-Instruct/c4a5702b42ed466de9e7.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"head_dim": 64,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 2048,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 8192,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "llama",
|
| 19 |
+
"neuron": {
|
| 20 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 21 |
+
"batch_size": 1,
|
| 22 |
+
"capacity_factor": null,
|
| 23 |
+
"checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
|
| 24 |
+
"checkpoint_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
|
| 25 |
+
"continuous_batching": false,
|
| 26 |
+
"ep_degree": 1,
|
| 27 |
+
"fused_qkv": true,
|
| 28 |
+
"glu_mlp": true,
|
| 29 |
+
"local_ranks_size": 2,
|
| 30 |
+
"max_batch_size": 1,
|
| 31 |
+
"max_context_length": 4096,
|
| 32 |
+
"max_topk": 256,
|
| 33 |
+
"n_active_tokens": 4096,
|
| 34 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 35 |
+
"on_device_sampling": true,
|
| 36 |
+
"optimum_neuron_version": "0.4.5.dev1",
|
| 37 |
+
"output_logits": false,
|
| 38 |
+
"pp_degree": 1,
|
| 39 |
+
"sequence_length": 4096,
|
| 40 |
+
"speculation_length": 0,
|
| 41 |
+
"start_rank_id": 0,
|
| 42 |
+
"target": "trn1",
|
| 43 |
+
"torch_dtype": "bfloat16",
|
| 44 |
+
"tp_degree": 2
|
| 45 |
+
},
|
| 46 |
+
"num_attention_heads": 32,
|
| 47 |
+
"num_hidden_layers": 16,
|
| 48 |
+
"num_key_value_heads": 8,
|
| 49 |
+
"pretraining_tp": 1,
|
| 50 |
+
"rms_norm_eps": 1e-05,
|
| 51 |
+
"rope_scaling": {
|
| 52 |
+
"factor": 32.0,
|
| 53 |
+
"high_freq_factor": 4.0,
|
| 54 |
+
"low_freq_factor": 1.0,
|
| 55 |
+
"original_max_position_embeddings": 8192,
|
| 56 |
+
"rope_type": "llama3"
|
| 57 |
+
},
|
| 58 |
+
"rope_theta": 500000.0,
|
| 59 |
+
"tie_word_embeddings": true,
|
| 60 |
+
"unsloth_fixed": true,
|
| 61 |
+
"use_cache": true,
|
| 62 |
+
"vocab_size": 128256
|
| 63 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/qwen2/Qwen/Qwen2.5-0.5B/df54a6f9b56ca3bdfc41.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 2,
|
| 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": 2,
|
| 55 |
+
"max_context_length": 1024,
|
| 56 |
+
"max_topk": 256,
|
| 57 |
+
"n_active_tokens": 1024,
|
| 58 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 59 |
+
"on_device_sampling": false,
|
| 60 |
+
"optimum_neuron_version": "0.4.5.dev1",
|
| 61 |
+
"output_logits": false,
|
| 62 |
+
"pp_degree": 1,
|
| 63 |
+
"sequence_length": 1024,
|
| 64 |
+
"speculation_length": 0,
|
| 65 |
+
"start_rank_id": 0,
|
| 66 |
+
"target": "trn1",
|
| 67 |
+
"torch_dtype": "bfloat16",
|
| 68 |
+
"tp_degree": 2
|
| 69 |
+
},
|
| 70 |
+
"num_attention_heads": 14,
|
| 71 |
+
"num_hidden_layers": 24,
|
| 72 |
+
"num_key_value_heads": 2,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_scaling": null,
|
| 75 |
+
"rope_theta": 1000000.0,
|
| 76 |
+
"sliding_window": null,
|
| 77 |
+
"tie_word_embeddings": true,
|
| 78 |
+
"use_cache": true,
|
| 79 |
+
"use_mrope": false,
|
| 80 |
+
"use_sliding_window": false,
|
| 81 |
+
"vocab_size": 151936
|
| 82 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/smollm3/HuggingFaceTB/SmolLM3-3B/1a86e1cfcaf229143d5c.json
ADDED
|
@@ -0,0 +1,134 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "HuggingFaceTB/SmolLM3-3B",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"SmolLM3ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 11008,
|
| 15 |
+
"layer_types": [
|
| 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 |
+
"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 |
+
],
|
| 53 |
+
"max_position_embeddings": 65536,
|
| 54 |
+
"max_window_layers": 28,
|
| 55 |
+
"mlp_bias": false,
|
| 56 |
+
"model_type": "smollm3",
|
| 57 |
+
"neuron": {
|
| 58 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 59 |
+
"batch_size": 1,
|
| 60 |
+
"capacity_factor": null,
|
| 61 |
+
"checkpoint_id": "HuggingFaceTB/SmolLM3-3B",
|
| 62 |
+
"checkpoint_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
|
| 63 |
+
"continuous_batching": false,
|
| 64 |
+
"ep_degree": 1,
|
| 65 |
+
"fused_qkv": true,
|
| 66 |
+
"glu_mlp": true,
|
| 67 |
+
"local_ranks_size": 2,
|
| 68 |
+
"max_batch_size": 1,
|
| 69 |
+
"max_context_length": 1024,
|
| 70 |
+
"max_topk": 256,
|
| 71 |
+
"n_active_tokens": 1024,
|
| 72 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 73 |
+
"on_device_sampling": true,
|
| 74 |
+
"optimum_neuron_version": "0.4.5.dev1",
|
| 75 |
+
"output_logits": false,
|
| 76 |
+
"pp_degree": 1,
|
| 77 |
+
"sequence_length": 1024,
|
| 78 |
+
"speculation_length": 0,
|
| 79 |
+
"start_rank_id": 0,
|
| 80 |
+
"target": "trn1",
|
| 81 |
+
"torch_dtype": "bfloat16",
|
| 82 |
+
"tp_degree": 2
|
| 83 |
+
},
|
| 84 |
+
"no_rope_layer_interval": 4,
|
| 85 |
+
"no_rope_layers": [
|
| 86 |
+
1,
|
| 87 |
+
1,
|
| 88 |
+
1,
|
| 89 |
+
0,
|
| 90 |
+
1,
|
| 91 |
+
1,
|
| 92 |
+
1,
|
| 93 |
+
0,
|
| 94 |
+
1,
|
| 95 |
+
1,
|
| 96 |
+
1,
|
| 97 |
+
0,
|
| 98 |
+
1,
|
| 99 |
+
1,
|
| 100 |
+
1,
|
| 101 |
+
0,
|
| 102 |
+
1,
|
| 103 |
+
1,
|
| 104 |
+
1,
|
| 105 |
+
0,
|
| 106 |
+
1,
|
| 107 |
+
1,
|
| 108 |
+
1,
|
| 109 |
+
0,
|
| 110 |
+
1,
|
| 111 |
+
1,
|
| 112 |
+
1,
|
| 113 |
+
0,
|
| 114 |
+
1,
|
| 115 |
+
1,
|
| 116 |
+
1,
|
| 117 |
+
0,
|
| 118 |
+
1,
|
| 119 |
+
1,
|
| 120 |
+
1,
|
| 121 |
+
0
|
| 122 |
+
],
|
| 123 |
+
"num_attention_heads": 16,
|
| 124 |
+
"num_hidden_layers": 36,
|
| 125 |
+
"num_key_value_heads": 4,
|
| 126 |
+
"pretraining_tp": 2,
|
| 127 |
+
"rms_norm_eps": 1e-06,
|
| 128 |
+
"rope_scaling": null,
|
| 129 |
+
"rope_theta": 5000000.0,
|
| 130 |
+
"sliding_window": null,
|
| 131 |
+
"use_cache": false,
|
| 132 |
+
"use_sliding_window": false,
|
| 133 |
+
"vocab_size": 128256
|
| 134 |
+
}
|
neuronxcc-2.21.33363.0+82129205/MODULE_10217061096959125489+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_10217061096959125489+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_10217061096959125489+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_10244305442015770634+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_10244305442015770634+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_10244305442015770634+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_10606948783918825529+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_10606948783918825529+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_10606948783918825529+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_10645643398657092095+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_10645643398657092095+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_10645643398657092095+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_10661660426924300837+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_10661660426924300837+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_10661660426924300837+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_10746122569655005679+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_10746122569655005679+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_10746122569655005679+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_11086318750207148626+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_11086318750207148626+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_11086318750207148626+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_11909310600244571805+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_11909310600244571805+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_11909310600244571805+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_11933818254123612383+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_11933818254123612383+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_11933818254123612383+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_1210392327607194823+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_1210392327607194823+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_1210392327607194823+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_12487216553200321032+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_12487216553200321032+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_12487216553200321032+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_13085549342645515693+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_13085549342645515693+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_13085549342645515693+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_13102010590082783346+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_13102010590082783346+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_13102010590082783346+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_13475273426270778455+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_13475273426270778455+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_13475273426270778455+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_13728813963059599796+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_13728813963059599796+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_13728813963059599796+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_13783872956654054643+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_13783872956654054643+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_13783872956654054643+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_14097776130612710282+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_14097776130612710282+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_14097776130612710282+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_14155105943057125178+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_14155105943057125178+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_14155105943057125178+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_14173682194645188821+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_14173682194645188821+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_14173682194645188821+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_14274833313744358348+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_14274833313744358348+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_14274833313744358348+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_14469716438804935215+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_14469716438804935215+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_14469716438804935215+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_14739672283771660808+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_14739672283771660808+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_14739672283771660808+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_15442663025941492357+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_15442663025941492357+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_15442663025941492357+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_15557971880365771457+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_15557971880365771457+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_15557971880365771457+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_15564425602929126510+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_15564425602929126510+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_15564425602929126510+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_15589838460896944293+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_15589838460896944293+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_15589838460896944293+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_15654572849171857535+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_15654572849171857535+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_15654572849171857535+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_15767383571209512795+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_15767383571209512795+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_15767383571209512795+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_15888836045088309511+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_15888836045088309511+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_15888836045088309511+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_15929451261464042997+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_15929451261464042997+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_15929451261464042997+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_16016505958416521648+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_16016505958416521648+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_16016505958416521648+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_16053163479112702088+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_16053163479112702088+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_16053163479112702088+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_16204387852795926216+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_16204387852795926216+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_16204387852795926216+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_1656719109221189948+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_1656719109221189948+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_1656719109221189948+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_16599571375348449904+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_16599571375348449904+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_16599571375348449904+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_1663757798483801648+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_1663757798483801648+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_1663757798483801648+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_16738296820980389103+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_16738296820980389103+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_16738296820980389103+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_1696825468766062114+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_1696825468766062114+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_1696825468766062114+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_17476209562158013765+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_17476209562158013765+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_17476209562158013765+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_17781596253725927902+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_17781596253725927902+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_17781596253725927902+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_1795740353831178306+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_1795740353831178306+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_1795740353831178306+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_1800832390737682969+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_1800832390737682969+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_1800832390737682969+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_1805481651134498710+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_1805481651134498710+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_1805481651134498710+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_1805728312057401221+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_1805728312057401221+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_1805728312057401221+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_18070749384531238695+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_18070749384531238695+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_18070749384531238695+e30acd3a/model.neff differ
|
|
|
neuronxcc-2.21.33363.0+82129205/MODULE_18134517905922687964+e30acd3a/model.neff
CHANGED
|
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_18134517905922687964+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_18134517905922687964+e30acd3a/model.neff differ
|
|
|