Add Neuron-optimized files for microsoft/Phi-3-mini-4k-instruct

🤖 Neuron Export Bot: Adding AWS Neuron-optimized model files.

Original model: [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct)
Task: text-generation
Generated by: [badaoui](https://huggingface.co/badaoui)
Generated using: [Optimum Neuron Compiler Space](https://huggingface.co/spaces/optimum/neuron-export)

These files have been pre-compiled for AWS Neuron devices (Inferentia/Trainium) and should provide improved inference performance.

README.md

@@ -0,0 +1,38 @@
+---
+tags:
+- neuron
+- optimized
+- aws-neuron
+- text-generation
+base_model: microsoft/Phi-3-mini-4k-instruct
+---
+
+# Neuron-Optimized microsoft/Phi-3-mini-4k-instruct
+
+This repository contains AWS Neuron-optimized files for [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct).
+
+## Model Details
+
+- **Base Model**: [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct)
+- **Task**: text-generation
+- **Optimization**: AWS Neuron compilation
+- **Generated by**: [badaoui](https://huggingface.co/badaoui)
+- **Generated using**: [Optimum Neuron Compiler Space](https://huggingface.co/spaces/optimum/neuron-export)
+
+## Usage
+
+This model has been optimized for AWS Neuron devices (Inferentia/Trainium). To use it:
+
+```python
+from optimum.neuron import NeuronModelForCausalLM
+
+model = NeuronModelForCausalLM.from_pretrained("badaoui/microsoft-Phi-3-mini-4k-instruct-neuron")
+```
+
+## Performance
+
+These files are pre-compiled for AWS Neuron devices and should provide improved inference performance compared to the original model when deployed on Inferentia or Trainium instances.
+
+## Original Model
+
+For the original model, training details, and more information, please visit: [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct)

config.json

@@ -0,0 +1,36 @@
+{
+  "architectures": [
+    "Phi3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "microsoft/Phi-3-mini-4k-instruct--configuration_phi3.Phi3Config",
+    "AutoModelForCausalLM": "microsoft/Phi-3-mini-4k-instruct--modeling_phi3.Phi3ForCausalLM"
+  },
+  "bos_token_id": 1,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 4096,
+  "model_type": "phi3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "original_max_position_embeddings": 4096,
+  "pad_token_id": 32000,
+  "partial_rotary_factor": 1.0,
+  "resid_pdrop": 0.0,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "sliding_window": 2047,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "vocab_size": 32064
+}

model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ea05eef0e85c15cb2a58b01c816b63ad3cb2517647442d9cf45e65014af850a
+size 8013148023

neuron_config.json

@@ -0,0 +1,43 @@
+{
+  "_serialized_key": "NxDNeuronConfig",
+  "async_mode": false,
+  "attn_kernel_enabled": false,
+  "batch_size": 1,
+  "capacity_factor": null,
+  "cc_pipeline_tiling_factor": 2,
+  "checkpoint_id": "microsoft/Phi-3-mini-4k-instruct",
+  "checkpoint_revision": "0a67737cc96d2554230f90338b163bc6380a2a85",
+  "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": 12,
+  "logical_nc_config": 1,
+  "max_batch_size": 1,
+  "max_context_length": 128,
+  "max_topk": 256,
+  "mlp_kernel_enabled": false,
+  "mlp_kernel_fuse_residual_add": false,
+  "n_active_tokens": 128,
+  "neuronxcc_version": "2.19.8089.0+8ab9f450",
+  "num_cores_per_group": 1,
+  "on_device_sampling": true,
+  "optimum_neuron_version": "0.3.0",
+  "output_logits": false,
+  "padding_side": "right",
+  "pp_degree": 1,
+  "qk_layernorm": false,
+  "qkv_kernel_enabled": false,
+  "rpl_reduce_dtype": "bfloat16",
+  "sequence_length": 128,
+  "sequence_parallel_enabled": false,
+  "speculation_length": 0,
+  "start_rank_id": 0,
+  "target": null,
+  "torch_dtype": "bfloat16",
+  "tp_degree": 12,
+  "vocab_parallel": false
+}