Synchronizing local compiler cache.

.gitattributes

@@ -17798,3 +17798,6 @@ neuronxcc-2.26.6360.0+6f180f47/MODULE_f75715e2ead36890b407+84337dd9/wrapped_neff
 neuronxcc-2.26.6360.0+6f180f47/MODULE_abc7f6e1bd3fc58a04bf+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
 neuronxcc-2.26.6360.0+6f180f47/MODULE_cfc98d6eaa0cefa6f7ca+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
 neuronxcc-2.26.6360.0+6f180f47/MODULE_cfc98d6eaa0cefa6f7ca+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.26.6360.0+6f180f47/MODULE_3a4f0acd9d5104d5fff2+8f3c2f7c/model.neff filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.26.6360.0+6f180f47/MODULE_3a4f0acd9d5104d5fff2+8f3c2f7c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.26.6360.0+6f180f47/MODULE_84423616ad3ab8d098f4+26701361/model.neff filter=lfs diff=lfs merge=lfs -text

neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/2184b8e912086f10a4fd3384fe48c361a358c591f0bc2d2020f1aee4498c3f22/d59ebd7bb9a3620e81a8.json

@@ -0,0 +1,59 @@
+{
+  "_entry_class": "SingleModelCacheEntry",
+  "_model_id": "mistralai/Mixtral-8x22B-Instruct-v0.1",
+  "_task": "text-generation",
+  "architectures": [
+    "MixtralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 6144,
+  "initializer_range": 0.02,
+  "intermediate_size": 16384,
+  "max_position_embeddings": 65536,
+  "model_type": "mixtral",
+  "neuron": {
+    "_serialized_key": "NxDNeuronConfig",
+    "batch_size": 4,
+    "capacity_factor": null,
+    "checkpoint_id": "mistralai/Mixtral-8x22B-Instruct-v0.1",
+    "checkpoint_revision": "cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d",
+    "continuous_batching": false,
+    "ep_degree": 1,
+    "fused_qkv": false,
+    "glu_mlp": true,
+    "local_ranks_size": 32,
+    "max_batch_size": 4,
+    "max_context_length": 4096,
+    "max_topk": 256,
+    "n_active_tokens": 4096,
+    "neuronxcc_version": "2.26.6360.0+6f180f47",
+    "on_device_sampling": false,
+    "optimum_neuron_version": "0.4.7.dev0",
+    "output_logits": false,
+    "pp_degree": 1,
+    "prefill_chunk_size": 1024,
+    "sequence_length": 4096,
+    "speculation_length": 0,
+    "start_rank_id": 0,
+    "target": "trn1",
+    "torch_dtype": "bfloat16",
+    "tp_degree": 32
+  },
+  "num_attention_heads": 48,
+  "num_experts_per_tok": 2,
+  "num_hidden_layers": 56,
+  "num_key_value_heads": 8,
+  "num_local_experts": 8,
+  "output_router_logits": false,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "router_aux_loss_coef": 0.001,
+  "router_jitter_noise": 0.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "use_cache": true,
+  "vocab_size": 32768
+}

neuronxcc-2.26.6360.0+6f180f47/MODULE_3a4f0acd9d5104d5fff2+8f3c2f7c/compile_flags.json

@@ -0,0 +1 @@
+["--target=inf2", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--verbose=35", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]

neuronxcc-2.26.6360.0+6f180f47/MODULE_3a4f0acd9d5104d5fff2+8f3c2f7c/model.hlo_module.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7048adb234199c2f581f97a2edc06b9412347a7fb984bb64ac89eec599c021f5
+size 10855190

neuronxcc-2.26.6360.0+6f180f47/MODULE_3a4f0acd9d5104d5fff2+8f3c2f7c/model.neff

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:963bfa8109ea8212a77f4bbe4132374499b465ef12418dfa5ea606ad1acf0c32
+size 11234304

neuronxcc-2.26.6360.0+6f180f47/MODULE_3a4f0acd9d5104d5fff2+8f3c2f7c/wrapped_neff.hlo

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb8eabc06d9dc98614e317e1fa874e2f882e3e825a1bdeee8b596c8f217f926b
+size 11562381

neuronxcc-2.26.6360.0+6f180f47/MODULE_84423616ad3ab8d098f4+26701361/compile_flags.json

@@ -0,0 +1 @@
+["--target=inf2", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--verbose=35", "--logfile=/tmp/nxd_model/chunked_prefill/_tp0_bk0/log-neuron-cc.txt"]

neuronxcc-2.26.6360.0+6f180f47/MODULE_84423616ad3ab8d098f4+26701361/model.hlo_module.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa2cbd718f0415c8ad7ac382286c4391c69fa27509b3b1de48f1022a2d8fd91d
+size 104620720

neuronxcc-2.26.6360.0+6f180f47/MODULE_84423616ad3ab8d098f4+26701361/model.neff

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8b3ba5f6b51e6ac3384b585f70b9295bbe314040c8c15c13612375d4b9e7e8c
+size 9585664