Synchronizing local compiler cache.
Browse files- .gitattributes +3 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/929b02754a13cbfdf657d863c3fc6f3bce672879bc6ae48ab45be21e881e9ec2/728acc8ab195811da6d8.json +88 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/qwen3/Qwen/Qwen3-0.6B/728acc8ab195811da6d8.json +88 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_5877d866ab9c5e16f0ce+a32116a7/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_5877d866ab9c5e16f0ce+a32116a7/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_5877d866ab9c5e16f0ce+a32116a7/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_5877d866ab9c5e16f0ce+a32116a7/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/wrapped_neff.hlo +3 -0
.gitattributes
CHANGED
|
@@ -7460,3 +7460,6 @@ neuronxcc-2.23.6484.0+3b612583/MODULE_6933000544295673404+fad94d7c/model.neff fi
|
|
| 7460 |
neuronxcc-2.21.33363.0+82129205/MODULE_370267033378699b0606+ac10809c/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7461 |
neuronxcc-2.21.33363.0+82129205/MODULE_370267033378699b0606+ac10809c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 7462 |
neuronxcc-2.21.33363.0+82129205/MODULE_f6541324218c275df145+593aa068/model.neff filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 7460 |
neuronxcc-2.21.33363.0+82129205/MODULE_370267033378699b0606+ac10809c/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7461 |
neuronxcc-2.21.33363.0+82129205/MODULE_370267033378699b0606+ac10809c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 7462 |
neuronxcc-2.21.33363.0+82129205/MODULE_f6541324218c275df145+593aa068/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7463 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_5877d866ab9c5e16f0ce+a32116a7/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7464 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7465 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/929b02754a13cbfdf657d863c3fc6f3bce672879bc6ae48ab45be21e881e9ec2/728acc8ab195811da6d8.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "Qwen/Qwen3-0.6B",
|
| 4 |
+
"_task": "text-generation",
|
| 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": 40960,
|
| 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/Qwen3-0.6B",
|
| 54 |
+
"checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
|
| 55 |
+
"continuous_batching": true,
|
| 56 |
+
"ep_degree": 1,
|
| 57 |
+
"fused_qkv": true,
|
| 58 |
+
"glu_mlp": true,
|
| 59 |
+
"local_ranks_size": 1,
|
| 60 |
+
"max_batch_size": 4,
|
| 61 |
+
"max_context_length": 1024,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 1024,
|
| 64 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 65 |
+
"on_device_sampling": true,
|
| 66 |
+
"optimum_neuron_version": "0.4.6.dev3",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"prefill_chunk_size": 0,
|
| 70 |
+
"sequence_length": 1024,
|
| 71 |
+
"speculation_length": 0,
|
| 72 |
+
"start_rank_id": 0,
|
| 73 |
+
"target": "trn2",
|
| 74 |
+
"torch_dtype": "bfloat16",
|
| 75 |
+
"tp_degree": 1
|
| 76 |
+
},
|
| 77 |
+
"num_attention_heads": 16,
|
| 78 |
+
"num_hidden_layers": 28,
|
| 79 |
+
"num_key_value_heads": 8,
|
| 80 |
+
"rms_norm_eps": 1e-06,
|
| 81 |
+
"rope_scaling": null,
|
| 82 |
+
"rope_theta": 1000000,
|
| 83 |
+
"sliding_window": null,
|
| 84 |
+
"tie_word_embeddings": true,
|
| 85 |
+
"use_cache": true,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vocab_size": 151936
|
| 88 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/qwen3/Qwen/Qwen3-0.6B/728acc8ab195811da6d8.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "Qwen/Qwen3-0.6B",
|
| 4 |
+
"_task": "text-generation",
|
| 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": 40960,
|
| 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/Qwen3-0.6B",
|
| 54 |
+
"checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
|
| 55 |
+
"continuous_batching": true,
|
| 56 |
+
"ep_degree": 1,
|
| 57 |
+
"fused_qkv": true,
|
| 58 |
+
"glu_mlp": true,
|
| 59 |
+
"local_ranks_size": 1,
|
| 60 |
+
"max_batch_size": 4,
|
| 61 |
+
"max_context_length": 1024,
|
| 62 |
+
"max_topk": 256,
|
| 63 |
+
"n_active_tokens": 1024,
|
| 64 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 65 |
+
"on_device_sampling": true,
|
| 66 |
+
"optimum_neuron_version": "0.4.6.dev3",
|
| 67 |
+
"output_logits": false,
|
| 68 |
+
"pp_degree": 1,
|
| 69 |
+
"prefill_chunk_size": 0,
|
| 70 |
+
"sequence_length": 1024,
|
| 71 |
+
"speculation_length": 0,
|
| 72 |
+
"start_rank_id": 0,
|
| 73 |
+
"target": "trn2",
|
| 74 |
+
"torch_dtype": "bfloat16",
|
| 75 |
+
"tp_degree": 1
|
| 76 |
+
},
|
| 77 |
+
"num_attention_heads": 16,
|
| 78 |
+
"num_hidden_layers": 28,
|
| 79 |
+
"num_key_value_heads": 8,
|
| 80 |
+
"rms_norm_eps": 1e-06,
|
| 81 |
+
"rope_scaling": null,
|
| 82 |
+
"rope_theta": 1000000,
|
| 83 |
+
"sliding_window": null,
|
| 84 |
+
"tie_word_embeddings": true,
|
| 85 |
+
"use_cache": true,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vocab_size": 151936
|
| 88 |
+
}
|
neuronxcc-2.21.33363.0+82129205/MODULE_5877d866ab9c5e16f0ce+a32116a7/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/context_encoding/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_5877d866ab9c5e16f0ce+a32116a7/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_5877d866ab9c5e16f0ce+a32116a7/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:807c79ddd2138c25a62c9923748b0314cfb66417dad6a3edd5052f7eb11a6639
|
| 3 |
+
size 2043341
|
neuronxcc-2.21.33363.0+82129205/MODULE_5877d866ab9c5e16f0ce+a32116a7/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d20d1e7bb78acad5027e89193733b391fbf2f7fbf49cd0488221d28240c7ba54
|
| 3 |
+
size 10630144
|
neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/compile_flags.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f0a4ec29321241ae8f6c99980f6134d8b18a4d3f725570202a2f25c696388bba
|
| 3 |
+
size 3698605
|
neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe8bdfdb41fb93791f827fcc5df269ceea7b9af092cb0122dbfb08484bf61d7e
|
| 3 |
+
size 7455744
|
neuronxcc-2.21.33363.0+82129205/MODULE_bd5774b51eeca5657757+ac10809c/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f783fe56879222d32f4feb2d9471d01f4bf0684a5e538b6c89679d28b90fed04
|
| 3 |
+
size 7593856
|