Synchronizing local compiler cache.

.gitattributes

@@ -8691,3 +8691,7 @@ neuronxcc-2.17.194.0+d312836f/MODULE_23cd41d05d4131cc4af0+431f5505/model.neff fi
 neuronxcc-2.17.194.0+d312836f/MODULE_b9df3a6b028991fa1f7f+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
 neuronxcc-2.17.194.0+d312836f/MODULE_b9df3a6b028991fa1f7f+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 neuronxcc-2.17.194.0+d312836f/MODULE_bd65a762e389c2a78a72+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.17.194.0+d312836f/MODULE_03b474270195263d85c9+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.17.194.0+d312836f/MODULE_375d9590fd5ac142c361+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.17.194.0+d312836f/MODULE_375d9590fd5ac142c361+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.17.194.0+d312836f/MODULE_398d98c0b8ee6124c193+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text

neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev2/granite/ibm-granite/granite-3.1-8b-instruct/2b1cec8d5e79797b25bb.json

@@ -0,0 +1,73 @@
+{
+  "_entry_class": "SingleModelCacheEntry",
+  "_model_id": "ibm-granite/granite-3.1-8b-instruct",
+  "_task": "text-generation",
+  "architectures": [
+    "GraniteForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.1,
+  "attention_multiplier": 0.0078125,
+  "embedding_multiplier": 12.0,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 12800,
+  "logits_scaling": 16.0,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "granite",
+  "neuron": {
+    "_serialized_key": "NxDNeuronConfig",
+    "async_mode": false,
+    "attn_kernel_enabled": false,
+    "batch_size": 1,
+    "capacity_factor": null,
+    "cc_pipeline_tiling_factor": 2,
+    "checkpoint_id": "ibm-granite/granite-3.1-8b-instruct",
+    "checkpoint_revision": "4009206d5fc95d2e65a7b7633e159d6e97e25d35",
+    "continuous_batching": false,
+    "enable_bucketing": false,
+    "ep_degree": 1,
+    "flash_decoding_enabled": false,
+    "fused_qkv": true,
+    "glu_mlp": true,
+    "is_chunked_prefill": false,
+    "local_ranks_size": 8,
+    "logical_nc_config": 1,
+    "max_batch_size": 1,
+    "max_context_length": 4096,
+    "max_topk": 256,
+    "mlp_kernel_enabled": false,
+    "mlp_kernel_fuse_residual_add": false,
+    "n_active_tokens": 4096,
+    "neuronxcc_version": "2.17.194.0+d312836f",
+    "num_cores_per_group": 1,
+    "on_device_sampling": true,
+    "optimum_neuron_version": "0.3.0.dev2",
+    "output_logits": false,
+    "padding_side": "right",
+    "pp_degree": 1,
+    "qk_layernorm": false,
+    "qkv_kernel_enabled": false,
+    "rpl_reduce_dtype": "bfloat16",
+    "sequence_length": 4096,
+    "sequence_parallel_enabled": false,
+    "speculation_length": 0,
+    "start_rank_id": 0,
+    "target": null,
+    "torch_dtype": "bfloat16",
+    "tp_degree": 8,
+    "vocab_parallel": false
+  },
+  "num_attention_heads": 32,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 8,
+  "residual_multiplier": 0.22,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000000.0,
+  "tie_word_embeddings": true,
+  "use_cache": true,
+  "vocab_size": 49155
+}

neuronxcc-2.17.194.0+d312836f/MODULE_03b474270195263d85c9+bfe5714b/compile_flags.json

@@ -0,0 +1 @@
+"--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2  --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"

neuronxcc-2.17.194.0+d312836f/MODULE_03b474270195263d85c9+bfe5714b/model.hlo_module.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e401868c47ec7c88fb89942d0b905c5598e53b7c2577f1f21a19f4a3e1feee3
+size 875353

neuronxcc-2.17.194.0+d312836f/MODULE_03b474270195263d85c9+bfe5714b/model.neff

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb62741edcb87ef99fed697ceb6d2f5718708ce5ef460651fc55c45a914685d1
+size 2264064

neuronxcc-2.17.194.0+d312836f/MODULE_375d9590fd5ac142c361+165e9558/compile_flags.json

@@ -0,0 +1 @@
+"--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2  --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt --enable-internal-neff-wrapper"

neuronxcc-2.17.194.0+d312836f/MODULE_375d9590fd5ac142c361+165e9558/model.hlo_module.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2d9ffcd124961e73db039d1f4793b7ae5a7a23957523e48aefa76da15634e72
+size 996569

neuronxcc-2.17.194.0+d312836f/MODULE_375d9590fd5ac142c361+165e9558/model.neff

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f52cfd63a8e3070cf913a931e22a815246d19a702db0f60a48f682cbc6c20f31
+size 2591744

neuronxcc-2.17.194.0+d312836f/MODULE_375d9590fd5ac142c361+165e9558/wrapped_neff.hlo

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24108632de599e918995248895ac2cc74a10dc34d193fa8baf20ed35d1cc0af6
+size 2763380

neuronxcc-2.17.194.0+d312836f/MODULE_398d98c0b8ee6124c193+431f5505/compile_flags.json

@@ -0,0 +1 @@
+"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt"

neuronxcc-2.17.194.0+d312836f/MODULE_398d98c0b8ee6124c193+431f5505/model.hlo_module.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea85b4239178fcc7e2081cbe0a9b6443c39e1c7f1792ce3103bc4f19215dbef0
+size 169362

neuronxcc-2.17.194.0+d312836f/MODULE_398d98c0b8ee6124c193+431f5505/model.neff

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:568dc0a3e5ad4923a375b38bfb1f596e5ea85ef4423dfec1e3e02a6ed242dbcf
+size 2243584