Synchronizing local compiler cache.
Browse files- .gitattributes +3 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/d915dcdd91e8e2f04ff262f036618c2a933bae11f6700cdeb4bb8d5c1aca1be0/cb17208de8e8a0db6ce6.json +87 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/qwen2/Qwen/Qwen2.5-1.5B/cb17208de8e8a0db6ce6.json +87 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_96d5253fce592e9d0389+a02c3a36/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_96d5253fce592e9d0389+a02c3a36/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_96d5253fce592e9d0389+a02c3a36/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_96d5253fce592e9d0389+a02c3a36/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_96d5253fce592e9d0389+a02c3a36/wrapped_neff.hlo +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_c68932cfa906e878e55b+24129607/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_c68932cfa906e878e55b+24129607/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_c68932cfa906e878e55b+24129607/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_c68932cfa906e878e55b+24129607/model.neff +3 -0
.gitattributes
CHANGED
|
@@ -7762,3 +7762,6 @@ neuronxcc-2.23.6484.0+3b612583/MODULE_e25122de1346508d5af6+781055f7/model.neff f
|
|
| 7762 |
neuronxcc-2.23.6484.0+3b612583/MODULE_644c1bd4f6c66322f9fa+781055f7/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7763 |
neuronxcc-2.23.6484.0+3b612583/MODULE_fb55f6c0bcc9f2e19b5c+dae43e99/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7764 |
neuronxcc-2.23.6484.0+3b612583/MODULE_fb55f6c0bcc9f2e19b5c+dae43e99/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 7762 |
neuronxcc-2.23.6484.0+3b612583/MODULE_644c1bd4f6c66322f9fa+781055f7/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7763 |
neuronxcc-2.23.6484.0+3b612583/MODULE_fb55f6c0bcc9f2e19b5c+dae43e99/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7764 |
neuronxcc-2.23.6484.0+3b612583/MODULE_fb55f6c0bcc9f2e19b5c+dae43e99/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 7765 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_96d5253fce592e9d0389+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 7766 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_96d5253fce592e9d0389+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 7767 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_c68932cfa906e878e55b+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/d915dcdd91e8e2f04ff262f036618c2a933bae11f6700cdeb4bb8d5c1aca1be0/cb17208de8e8a0db6ce6.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "Qwen/Qwen2.5-1.5B",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen2ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 1536,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 8960,
|
| 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 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention"
|
| 43 |
+
],
|
| 44 |
+
"max_position_embeddings": 131072,
|
| 45 |
+
"max_window_layers": 28,
|
| 46 |
+
"model_type": "qwen2",
|
| 47 |
+
"neuron": {
|
| 48 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 49 |
+
"batch_size": 4,
|
| 50 |
+
"capacity_factor": null,
|
| 51 |
+
"checkpoint_id": "Qwen/Qwen2.5-1.5B",
|
| 52 |
+
"checkpoint_revision": "8faed761d45a263340a0528343f099c05c9a4323",
|
| 53 |
+
"continuous_batching": true,
|
| 54 |
+
"ep_degree": 1,
|
| 55 |
+
"fused_qkv": false,
|
| 56 |
+
"glu_mlp": true,
|
| 57 |
+
"local_ranks_size": 2,
|
| 58 |
+
"max_batch_size": 4,
|
| 59 |
+
"max_context_length": 1024,
|
| 60 |
+
"max_topk": 256,
|
| 61 |
+
"n_active_tokens": 1024,
|
| 62 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 63 |
+
"on_device_sampling": false,
|
| 64 |
+
"optimum_neuron_version": "0.4.6.dev3",
|
| 65 |
+
"output_logits": false,
|
| 66 |
+
"pp_degree": 1,
|
| 67 |
+
"prefill_chunk_size": 0,
|
| 68 |
+
"sequence_length": 1024,
|
| 69 |
+
"speculation_length": 0,
|
| 70 |
+
"start_rank_id": 0,
|
| 71 |
+
"target": "trn1",
|
| 72 |
+
"torch_dtype": "bfloat16",
|
| 73 |
+
"tp_degree": 2
|
| 74 |
+
},
|
| 75 |
+
"num_attention_heads": 12,
|
| 76 |
+
"num_hidden_layers": 28,
|
| 77 |
+
"num_key_value_heads": 2,
|
| 78 |
+
"rms_norm_eps": 1e-06,
|
| 79 |
+
"rope_scaling": null,
|
| 80 |
+
"rope_theta": 1000000.0,
|
| 81 |
+
"sliding_window": null,
|
| 82 |
+
"tie_word_embeddings": true,
|
| 83 |
+
"use_cache": true,
|
| 84 |
+
"use_mrope": false,
|
| 85 |
+
"use_sliding_window": false,
|
| 86 |
+
"vocab_size": 151936
|
| 87 |
+
}
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/qwen2/Qwen/Qwen2.5-1.5B/cb17208de8e8a0db6ce6.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "Qwen/Qwen2.5-1.5B",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen2ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 1536,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 8960,
|
| 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 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention"
|
| 43 |
+
],
|
| 44 |
+
"max_position_embeddings": 131072,
|
| 45 |
+
"max_window_layers": 28,
|
| 46 |
+
"model_type": "qwen2",
|
| 47 |
+
"neuron": {
|
| 48 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 49 |
+
"batch_size": 4,
|
| 50 |
+
"capacity_factor": null,
|
| 51 |
+
"checkpoint_id": "Qwen/Qwen2.5-1.5B",
|
| 52 |
+
"checkpoint_revision": "8faed761d45a263340a0528343f099c05c9a4323",
|
| 53 |
+
"continuous_batching": true,
|
| 54 |
+
"ep_degree": 1,
|
| 55 |
+
"fused_qkv": false,
|
| 56 |
+
"glu_mlp": true,
|
| 57 |
+
"local_ranks_size": 2,
|
| 58 |
+
"max_batch_size": 4,
|
| 59 |
+
"max_context_length": 1024,
|
| 60 |
+
"max_topk": 256,
|
| 61 |
+
"n_active_tokens": 1024,
|
| 62 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 63 |
+
"on_device_sampling": false,
|
| 64 |
+
"optimum_neuron_version": "0.4.6.dev3",
|
| 65 |
+
"output_logits": false,
|
| 66 |
+
"pp_degree": 1,
|
| 67 |
+
"prefill_chunk_size": 0,
|
| 68 |
+
"sequence_length": 1024,
|
| 69 |
+
"speculation_length": 0,
|
| 70 |
+
"start_rank_id": 0,
|
| 71 |
+
"target": "trn1",
|
| 72 |
+
"torch_dtype": "bfloat16",
|
| 73 |
+
"tp_degree": 2
|
| 74 |
+
},
|
| 75 |
+
"num_attention_heads": 12,
|
| 76 |
+
"num_hidden_layers": 28,
|
| 77 |
+
"num_key_value_heads": 2,
|
| 78 |
+
"rms_norm_eps": 1e-06,
|
| 79 |
+
"rope_scaling": null,
|
| 80 |
+
"rope_theta": 1000000.0,
|
| 81 |
+
"sliding_window": null,
|
| 82 |
+
"tie_word_embeddings": true,
|
| 83 |
+
"use_cache": true,
|
| 84 |
+
"use_mrope": false,
|
| 85 |
+
"use_sliding_window": false,
|
| 86 |
+
"vocab_size": 151936
|
| 87 |
+
}
|
neuronxcc-2.21.33363.0+82129205/MODULE_96d5253fce592e9d0389+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_96d5253fce592e9d0389+a02c3a36/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_96d5253fce592e9d0389+a02c3a36/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b1bb986c0879e39990c7cce4352521e5573eadbdf4cc685ed07ff8249dfafbd
|
| 3 |
+
size 723640
|
neuronxcc-2.21.33363.0+82129205/MODULE_96d5253fce592e9d0389+a02c3a36/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52cdd5dd327dd5fbe6d26c4a699e3c328d54d56d0508d0eace7c628632587105
|
| 3 |
+
size 3124224
|
neuronxcc-2.21.33363.0+82129205/MODULE_96d5253fce592e9d0389+a02c3a36/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6958ed4ae8bc4d3071808fdc4059d87d20aef952b846f2cbd39eb745928050ae
|
| 3 |
+
size 3308285
|
neuronxcc-2.21.33363.0+82129205/MODULE_c68932cfa906e878e55b+24129607/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/context_encoding/_tp0_bk0/log-neuron-cc.txt"]
|
neuronxcc-2.21.33363.0+82129205/MODULE_c68932cfa906e878e55b+24129607/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_c68932cfa906e878e55b+24129607/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:306cba2c35c5cba89b87a24aef6d3dbc0f2d1c8db79485d441c584833a8e3135
|
| 3 |
+
size 696209
|
neuronxcc-2.21.33363.0+82129205/MODULE_c68932cfa906e878e55b+24129607/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:549fd12d535cb4c5ef01729c805950da3bc0be7cabcc79a0fac35b2caf255a16
|
| 3 |
+
size 7035904
|