Synchronizing local compiler cache.
Browse files- .gitattributes +3 -0
- neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.3.dev2/llama/meta-llama/Llama-3.2-1B/d0d6cfee45051f746bdf.json +62 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_27143ff5b0e3602becba+24129607/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_27143ff5b0e3602becba+24129607/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_27143ff5b0e3602becba+24129607/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_27143ff5b0e3602becba+24129607/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_92667f83a1c29e318728+a02c3a36/compile_flags.json +1 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_92667f83a1c29e318728+a02c3a36/model.done +0 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_92667f83a1c29e318728+a02c3a36/model.hlo_module.pb +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_92667f83a1c29e318728+a02c3a36/model.neff +3 -0
- neuronxcc-2.21.33363.0+82129205/MODULE_92667f83a1c29e318728+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes
CHANGED
|
@@ -13703,3 +13703,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_b829c6e3567ae63cc8cd+e0c92c88/model.neff
|
|
| 13703 |
neuronxcc-2.21.33363.0+82129205/MODULE_d4d0d8cf4c8987b96c75+a32116a7/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 13704 |
neuronxcc-2.21.33363.0+82129205/MODULE_ea3a007bc363c7595c33+ac10809c/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 13705 |
neuronxcc-2.21.33363.0+82129205/MODULE_ea3a007bc363c7595c33+ac10809c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 13703 |
neuronxcc-2.21.33363.0+82129205/MODULE_d4d0d8cf4c8987b96c75+a32116a7/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 13704 |
neuronxcc-2.21.33363.0+82129205/MODULE_ea3a007bc363c7595c33+ac10809c/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 13705 |
neuronxcc-2.21.33363.0+82129205/MODULE_ea3a007bc363c7595c33+ac10809c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
| 13706 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_27143ff5b0e3602becba+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 13707 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_92667f83a1c29e318728+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
|
| 13708 |
+
neuronxcc-2.21.33363.0+82129205/MODULE_92667f83a1c29e318728+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
|
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.3.dev2/llama/meta-llama/Llama-3.2-1B/d0d6cfee45051f746bdf.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "meta-llama/Llama-3.2-1B",
|
| 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": "meta-llama/Llama-3.2-1B",
|
| 24 |
+
"checkpoint_revision": "4e20de362430cd3b72f300e6b0f18e50e7166e08",
|
| 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": "0.4.3.dev2",
|
| 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 |
+
"use_cache": true,
|
| 61 |
+
"vocab_size": 128256
|
| 62 |
+
}
|
neuronxcc-2.21.33363.0+82129205/MODULE_27143ff5b0e3602becba+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_27143ff5b0e3602becba+24129607/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_27143ff5b0e3602becba+24129607/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52087e84b4ff7caca484fddd14a99801401a845edbf8cd8671b861697587d922
|
| 3 |
+
size 474402
|
neuronxcc-2.21.33363.0+82129205/MODULE_27143ff5b0e3602becba+24129607/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd9ae26fc5e32369b36cccda809a762cb0fe98bb45f8f238213b7af93dc62ca2
|
| 3 |
+
size 42322944
|
neuronxcc-2.21.33363.0+82129205/MODULE_92667f83a1c29e318728+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_92667f83a1c29e318728+a02c3a36/model.done
ADDED
|
File without changes
|
neuronxcc-2.21.33363.0+82129205/MODULE_92667f83a1c29e318728+a02c3a36/model.hlo_module.pb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e79e0864d9a0cb8832a5f87390ab3fde012b08a479e8a2e6c55b1d74844efb1
|
| 3 |
+
size 417652
|
neuronxcc-2.21.33363.0+82129205/MODULE_92667f83a1c29e318728+a02c3a36/model.neff
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb747c4dc945585f79ae175cf6663d69e54c22049e2dc5c683745003e49307d9
|
| 3 |
+
size 2120704
|
neuronxcc-2.21.33363.0+82129205/MODULE_92667f83a1c29e318728+a02c3a36/wrapped_neff.hlo
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68801df7b2c9f01c8ad6efdcffe519174a47862b18f8e074762a1e8d17bc5b3e
|
| 3 |
+
size 2194934
|