Synchronizing local compiler cache.
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +14 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/60feecaa0c4c075e2f3e46a3f55d9a273f0ddd75a0ecf64e4ae27352e0819506/5b290bca6bdaf6e38333.json +87 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/60feecaa0c4c075e2f3e46a3f55d9a273f0ddd75a0ecf64e4ae27352e0819506/c4c6aab3c906d2d39928.json +87 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/qwen3/Qwen/Qwen3-Embedding-0.6B/c4c6aab3c906d2d39928.json +87 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/wrapped_neff.hlo +3 -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
.gitattributes
CHANGED
|
@@ -15938,3 +15938,17 @@ neuronxcc-2.21.33363.0+82129205/MODULE_a5bd4c82318d7b6dfffc+e4cf080e/model.neff
|
|
| 15938 |
neuronxcc-2.21.33363.0+82129205/MODULE_a5bd4c82318d7b6dfffc+e4cf080e/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15939 |
neuronxcc-2.21.33363.0+82129205/MODULE_afcad8be9ea16e076401+005f226c/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15940 |
neuronxcc-2.21.33363.0+82129205/MODULE_afcad8be9ea16e076401+005f226c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15938 |
neuronxcc-2.21.33363.0+82129205/MODULE_a5bd4c82318d7b6dfffc+e4cf080e/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15939 |
neuronxcc-2.21.33363.0+82129205/MODULE_afcad8be9ea16e076401+005f226c/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15940 |
neuronxcc-2.21.33363.0+82129205/MODULE_afcad8be9ea16e076401+005f226c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15941 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15942 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15943 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_235c514556fa231a940c+df7b731b/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15944 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_235c514556fa231a940c+df7b731b/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15945 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_2da63caa31e7595bc07f+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15946 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_2e5b91a5009bbf2b0b32+8dfc7133/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15947 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_2e5b91a5009bbf2b0b32+8dfc7133/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15948 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_7a217d914c9153d2c1e8+3afc4c26/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15949 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_7a217d914c9153d2c1e8+3afc4c26/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15950 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_88190e1b8a8ceb313e53+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15951 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_8f6a41a7be8028ed3e7f+ad2a59a7/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15952 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_8f6a41a7be8028ed3e7f+ad2a59a7/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15953 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_99a48037f942ccd7b85f+1912433e/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15954 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_99a48037f942ccd7b85f+1912433e/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/60feecaa0c4c075e2f3e46a3f55d9a273f0ddd75a0ecf64e4ae27352e0819506/5b290bca6bdaf6e38333.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "Qwen/Qwen3-Embedding-0.6B",
|
| 4 |
+
"_task": "feature-extraction",
|
| 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": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 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": 32768,
|
| 47 |
+
"max_window_layers": 28,
|
| 48 |
+
"model_type": "qwen3",
|
| 49 |
+
"neuron": {
|
| 50 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 51 |
+
"batch_size": 6,
|
| 52 |
+
"capacity_factor": null,
|
| 53 |
+
"checkpoint_id": "Qwen/Qwen3-Embedding-0.6B",
|
| 54 |
+
"checkpoint_revision": "c54f2e6e80b2d7b7de06f51cec4959f6b3e03418",
|
| 55 |
+
"continuous_batching": false,
|
| 56 |
+
"ep_degree": 1,
|
| 57 |
+
"fused_qkv": true,
|
| 58 |
+
"glu_mlp": true,
|
| 59 |
+
"local_ranks_size": 2,
|
| 60 |
+
"max_batch_size": 6,
|
| 61 |
+
"max_context_length": 8192,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 8192,
|
| 64 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 65 |
+
"on_device_sampling": false,
|
| 66 |
+
"optimum_neuron_version": "0.4.5.dev1",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"sequence_length": 8192,
|
| 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": 151669
|
| 87 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/60feecaa0c4c075e2f3e46a3f55d9a273f0ddd75a0ecf64e4ae27352e0819506/c4c6aab3c906d2d39928.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "Qwen/Qwen3-Embedding-0.6B",
|
| 4 |
+
"_task": "feature-extraction",
|
| 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": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 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": 32768,
|
| 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-Embedding-0.6B",
|
| 54 |
+
"checkpoint_revision": "c54f2e6e80b2d7b7de06f51cec4959f6b3e03418",
|
| 55 |
+
"continuous_batching": false,
|
| 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": 8192,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 8192,
|
| 64 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 65 |
+
"on_device_sampling": false,
|
| 66 |
+
"optimum_neuron_version": "0.4.5.dev1",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"sequence_length": 8192,
|
| 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": 151669
|
| 87 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/qwen3/Qwen/Qwen3-Embedding-0.6B/c4c6aab3c906d2d39928.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "Qwen/Qwen3-Embedding-0.6B",
|
| 4 |
+
"_task": "feature-extraction",
|
| 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": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 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": 32768,
|
| 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-Embedding-0.6B",
|
| 54 |
+
"checkpoint_revision": "c54f2e6e80b2d7b7de06f51cec4959f6b3e03418",
|
| 55 |
+
"continuous_batching": false,
|
| 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": 8192,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 8192,
|
| 64 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 65 |
+
"on_device_sampling": false,
|
| 66 |
+
"optimum_neuron_version": "0.4.5.dev1",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"sequence_length": 8192,
|
| 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": 151669
|
| 87 |
+
}
|
neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn1", "--enable-saturate-infinity", "--auto-cast=none", "--model-type=transformer", "-O1", "--logfile=/tmp/nxdi_test_aa5f6840-bfc8-4870-beb5-2ed3394a8e8e/compiler_workdir/AttentionModelWrapper/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51e79a5cc325f9a317d158d931d75ec73ae1cfb16a4d83343a2593edd95d0044
|
| 3 |
+
size 8979
|
neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05b55744d09e461ea51ccc1da9874dcefac30c3b3a3685f334d422788c747817
|
| 3 |
+
size 246784
|
neuronxcc-2.21.33363.0+82129205/MODULE_04ab7c96e6ea9a5e1e9d+6fcbf226/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3504e5338a6f170e31cccf2081054a439f34796cc6d8beb87ac9378e2303304
|
| 3 |
+
size 249608
|
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
|
|
|