Synchronizing local compiler cache.
Browse files- .gitattributes +3 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/phi3/microsoft/Phi-3-mini-4k-instruct/6d3af12941fb6112d4a0.json +62 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_05b4056d3722a8b5ac63+24129607/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_05b4056d3722a8b5ac63+24129607/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_05b4056d3722a8b5ac63+24129607/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_05b4056d3722a8b5ac63+24129607/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_ee90b8d72dc7e09526c1+a02c3a36/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_ee90b8d72dc7e09526c1+a02c3a36/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_ee90b8d72dc7e09526c1+a02c3a36/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_ee90b8d72dc7e09526c1+a02c3a36/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_ee90b8d72dc7e09526c1+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes
CHANGED
|
@@ -15223,3 +15223,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_73d5b1e273c6e99941d5+ac10809c/wrapped_nef
|
|
| 15223 |
neuronxcc-2.21.33363.0+82129205/MODULE_5543c5de678b9313e648+80d05c3f/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15224 |
neuronxcc-2.21.33363.0+82129205/MODULE_5543c5de678b9313e648+80d05c3f/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15225 |
neuronxcc-2.21.33363.0+82129205/MODULE_c9bee89b9f27583adb86+6e4949b4/model.neff filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 15223 |
neuronxcc-2.21.33363.0+82129205/MODULE_5543c5de678b9313e648+80d05c3f/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15224 |
neuronxcc-2.21.33363.0+82129205/MODULE_5543c5de678b9313e648+80d05c3f/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 15225 |
neuronxcc-2.21.33363.0+82129205/MODULE_c9bee89b9f27583adb86+6e4949b4/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15226 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_05b4056d3722a8b5ac63+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15227 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_ee90b8d72dc7e09526c1+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 15228 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_ee90b8d72dc7e09526c1+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/phi3/microsoft/Phi-3-mini-4k-instruct/6d3af12941fb6112d4a0.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "microsoft/Phi-3-mini-4k-instruct",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Phi3ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"auto_map": {
|
| 11 |
+
"AutoConfig": "configuration_phi3.Phi3Config",
|
| 12 |
+
"AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
|
| 13 |
+
},
|
| 14 |
+
"dtype": "bfloat16",
|
| 15 |
+
"embd_pdrop": 0.0,
|
| 16 |
+
"hidden_act": "silu",
|
| 17 |
+
"hidden_size": 3072,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 8192,
|
| 20 |
+
"max_position_embeddings": 4096,
|
| 21 |
+
"model_type": "phi3",
|
| 22 |
+
"neuron": {
|
| 23 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 24 |
+
"batch_size": 4,
|
| 25 |
+
"capacity_factor": null,
|
| 26 |
+
"checkpoint_id": "microsoft/Phi-3-mini-4k-instruct",
|
| 27 |
+
"checkpoint_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
|
| 28 |
+
"continuous_batching": true,
|
| 29 |
+
"ep_degree": 1,
|
| 30 |
+
"fused_qkv": true,
|
| 31 |
+
"glu_mlp": true,
|
| 32 |
+
"local_ranks_size": 2,
|
| 33 |
+
"max_batch_size": 4,
|
| 34 |
+
"max_context_length": 4096,
|
| 35 |
+
"max_topk": 256,
|
| 36 |
+
"n_active_tokens": 4096,
|
| 37 |
+
"neuronxcc_version": "2.21.33363.0+82129205",
|
| 38 |
+
"on_device_sampling": true,
|
| 39 |
+
"optimum_neuron_version": "0.4.4.dev2",
|
| 40 |
+
"output_logits": false,
|
| 41 |
+
"pp_degree": 1,
|
| 42 |
+
"sequence_length": 4096,
|
| 43 |
+
"speculation_length": 0,
|
| 44 |
+
"start_rank_id": 0,
|
| 45 |
+
"target": "trn1",
|
| 46 |
+
"torch_dtype": "bfloat16",
|
| 47 |
+
"tp_degree": 2
|
| 48 |
+
},
|
| 49 |
+
"num_attention_heads": 32,
|
| 50 |
+
"num_hidden_layers": 32,
|
| 51 |
+
"num_key_value_heads": 32,
|
| 52 |
+
"original_max_position_embeddings": 4096,
|
| 53 |
+
"partial_rotary_factor": 1.0,
|
| 54 |
+
"resid_pdrop": 0.0,
|
| 55 |
+
"rms_norm_eps": 1e-05,
|
| 56 |
+
"rope_scaling": null,
|
| 57 |
+
"rope_theta": 10000.0,
|
| 58 |
+
"sliding_window": 2047,
|
| 59 |
+
"tie_word_embeddings": false,
|
| 60 |
+
"use_cache": true,
|
| 61 |
+
"vocab_size": 32064
|
| 62 |
+
}
|
neuronxcc-2.21.33363.0+82129205/MODULE_05b4056d3722a8b5ac63+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_05b4056d3722a8b5ac63+24129607/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_05b4056d3722a8b5ac63+24129607/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fed10f176b9bca042f5c4020394797dcfb4a5f35026a42456546ebf471d49535
|
| 3 |
+
size 923430
|
neuronxcc-2.21.33363.0+82129205/MODULE_05b4056d3722a8b5ac63+24129607/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:462bfcfbba80ebe3c1b4853afc48f48732b3203cb7e8f2d3becd3925a19091bc
|
| 3 |
+
size 5213184
|
neuronxcc-2.21.33363.0+82129205/MODULE_ee90b8d72dc7e09526c1+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_ee90b8d72dc7e09526c1+a02c3a36/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_ee90b8d72dc7e09526c1+a02c3a36/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:484795c355cd757b363e97908e9bf5d104a45bb99ddadd9fa6c5852dfded2b7d
|
| 3 |
+
size 700208
|
neuronxcc-2.21.33363.0+82129205/MODULE_ee90b8d72dc7e09526c1+a02c3a36/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3d5d556ecadcfc2b9b56fed04aea7d8cb63f70be2eef202c4e4f533a5a9fe26
|
| 3 |
+
size 12411904
|
neuronxcc-2.21.33363.0+82129205/MODULE_ee90b8d72dc7e09526c1+a02c3a36/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52791069352af6e2e252feb67f8b68648acadedc6d31eda337098806fc090179
|
| 3 |
+
size 12558920
|