Upload fine-tuned model with vector memory layer

added_tokens.json

@@ -0,0 +1,13 @@
+{
+  "<|assistant|>": 32001,
+  "<|endoftext|>": 32000,
+  "<|end|>": 32007,
+  "<|placeholder1|>": 32002,
+  "<|placeholder2|>": 32003,
+  "<|placeholder3|>": 32004,
+  "<|placeholder4|>": 32005,
+  "<|placeholder5|>": 32008,
+  "<|placeholder6|>": 32009,
+  "<|system|>": 32006,
+  "<|user|>": 32010
+}

architecture.py

@@ -0,0 +1,144 @@
+# --- START OF FILE architecture.py ---
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from transformers import Phi3Config, Phi3ForCausalLM
+from typing import Optional, Dict
+
+# --- BUILDING BLOCK 1: VectorMemoryHead (No changes needed here, it inherits dtype correctly) ---
+class VectorMemoryHead(nn.Module):
+    def __init__(self, hidden_dim: int, num_memory_slots: int, num_heads: int, ff_dim: int, device=None, dtype=None):
+        super().__init__()
+        self.hidden_dim = hidden_dim
+        self.num_memory_slots = num_memory_slots
+        encoder_layer = nn.TransformerEncoderLayer(
+            d_model=hidden_dim, nhead=num_heads, dim_feedforward=ff_dim, dropout=0.1, batch_first=True,
+            device=device, dtype=dtype
+        )
+        self.encoder = nn.TransformerEncoder(encoder_layer, num_layers=1)
+        self.memory_queries = nn.Parameter(torch.randn(1, num_memory_slots, hidden_dim, device=device, dtype=dtype))
+        self.memory_attention = nn.MultiheadAttention(
+            embed_dim=hidden_dim, num_heads=num_heads, dropout=0.1, batch_first=True,
+            device=device, dtype=dtype
+        )
+        self.memory_layernorm = nn.LayerNorm(hidden_dim, device=device, dtype=dtype)
+        self.decoder_attention = nn.MultiheadAttention(
+            embed_dim=hidden_dim, num_heads=num_heads, dropout=0.1, batch_first=True,
+            device=device, dtype=dtype
+        )
+        self.decoder_layernorm = nn.LayerNorm(hidden_dim, device=device, dtype=dtype)
+        self.decoder_ffn = nn.Sequential(
+            nn.Linear(hidden_dim, ff_dim, device=device, dtype=dtype),
+            nn.ReLU(),
+            nn.Linear(ff_dim, hidden_dim, device=device, dtype=dtype)
+        )
+
+    def forward(self, memory_input_sequence: torch.Tensor):
+        batch_size = memory_input_sequence.shape[0]
+        encoded_vectors = self.encoder(memory_input_sequence)
+        queries = self.memory_queries.expand(batch_size, -1, -1)
+        compressed_memory, _ = self.memory_attention(query=queries, key=encoded_vectors, value=encoded_vectors)
+        compressed_memory = self.memory_layernorm(compressed_memory + queries)
+        reconstructed, _ = self.decoder_attention(query=encoded_vectors, key=compressed_memory, value=compressed_memory)
+        reconstructed_vectors = self.decoder_layernorm(reconstructed + encoded_vectors)
+        reconstructed_vectors = self.decoder_ffn(reconstructed_vectors)
+        return compressed_memory, reconstructed_vectors
+
+# --- BUILDING BLOCK 2: The Custom Layer (With Iterative Self-Correction) ---
+class GCVectorMemoryLayer(nn.Module):
+    def __init__(self, original_layer: nn.Linear, global_input_dim: int,
+                 memory_dim: int, num_memory_slots: int, memory_num_heads: int,
+                 global_state_storage: Dict):
+        super().__init__()
+        self.input_dim = original_layer.in_features
+        self.output_dim = original_layer.out_features
+        self.memory_dim = memory_dim
+        self.global_state_storage = global_state_storage
+        self.linear = original_layer
+
+        device, dtype = self.linear.weight.device, self.linear.weight.dtype
+
+        # This part is correct: initialize with the correct dtype
+        self.local_state_proj = nn.Linear(self.input_dim, memory_dim, device=device, dtype=dtype)
+        self.global_state_proj = nn.Linear(global_input_dim, memory_dim, device=device, dtype=dtype)
+        self.memory_head = VectorMemoryHead(
+            hidden_dim=memory_dim, num_memory_slots=num_memory_slots,
+            num_heads=memory_num_heads, ff_dim=memory_dim * 2, device=device, dtype=dtype
+        )
+        self.correction_head = nn.Linear(memory_dim, 2 * self.output_dim, device=device, dtype=dtype)
+
+        # --- NEW: Parameter for iterative self-correction ---
+        # This can be changed at inference time to apply the correction multiple times.
+        # Default is 1 to match training behavior.
+        self.num_correction_passes: int = 1
+
+        self.last_corrected_activation: Optional[torch.Tensor] = None
+        self.last_additive_correction: Optional[torch.Tensor] = None
+        self.last_memory_input: Optional[torch.Tensor] = None
+        self.last_reconstructed_from_memory: Optional[torch.Tensor] = None
+
+    def forward(self, x: torch.Tensor):
+        base_output = self.linear(x)
+
+        # If no global state is available or correction is disabled, return base output.
+        if 'embeds' not in self.global_state_storage or self.num_correction_passes < 1:
+            return base_output
+
+        global_embeds = self.global_state_storage['embeds']
+        if global_embeds.shape[1] != x.shape[1]: global_embeds = global_embeds[:, -x.shape[1]:, :]
+        B, S, _ = x.shape
+
+        with torch.enable_grad():
+            # --- 1. Calculate the correction signal ONCE ---
+            proj_local = self.local_state_proj(x)
+            proj_global = self.global_state_proj(global_embeds)
+
+            memory_input = torch.stack([proj_global, proj_local], dim=2)
+            memory_input_flat = memory_input.view(B * S, 2, self.memory_dim)
+            compressed_mem_flat, recon_flat = self.memory_head(memory_input_flat)
+            aggregated_thought_flat = compressed_mem_flat.mean(dim=1)
+            aggregated_thought = aggregated_thought_flat.view(B, S, self.memory_dim)
+            raw_correction = self.correction_head(aggregated_thought)
+            gate, value = torch.chunk(raw_correction, 2, dim=-1)
+
+            # --- 2. Iteratively apply the correction signal ---
+            corrected_activation = base_output
+            for _ in range(self.num_correction_passes):
+                corrected_activation = corrected_activation * torch.sigmoid(gate.to(x.dtype)) + value.to(x.dtype)
+
+        # During training, store the final activation and the original correction signal
+        # for loss calculation.
+        if self.training:
+            self.last_corrected_activation = corrected_activation
+            self.last_additive_correction = value # The 'value' is the core additive signal
+            self.last_memory_input = memory_input_flat
+            self.last_reconstructed_from_memory = recon_flat
+
+        return corrected_activation
+
+# --- BUILDING BLOCK 3: The Full Custom Model ---
+class Phi3WithVectorMemoryForCausalLM(Phi3ForCausalLM):
+    def __init__(self, config):
+        super().__init__(config)
+        self.global_state_storage = {}
+        self.target_layer_path = "model.layers.15.mlp.gate_up_proj"
+
+        self.model.embed_tokens.register_forward_hook(
+            lambda module, input, output: self.global_state_storage.update({'embeds': output.detach()})
+        )
+
+        try:
+            original_layer = self.get_submodule(self.target_layer_path)
+            custom_layer = GCVectorMemoryLayer(
+                original_layer=original_layer, global_input_dim=config.hidden_size,
+                memory_dim=64, num_memory_slots=8, memory_num_heads=4,
+                global_state_storage=self.global_state_storage
+            )
+            parent_path = ".".join(self.target_layer_path.split('.')[:-1])
+            child_name = self.target_layer_path.split('.')[-1]
+            setattr(self.get_submodule(parent_path), child_name, custom_layer)
+            print(f"Successfully replaced '{self.target_layer_path}' with GCVectorMemoryLayer.")
+        except AttributeError:
+            print(f"Could not find target layer '{self.target_layer_path}'. Model remains unmodified.")
+# --- END OF FILE architecture.py ---

chat_template.jinja

@@ -0,0 +1,8 @@
+{% for message in messages %}{% if message['role'] == 'system' %}{{'<|system|>
+' + message['content'] + '<|end|>
+'}}{% elif message['role'] == 'user' %}{{'<|user|>
+' + message['content'] + '<|end|>
+'}}{% elif message['role'] == 'assistant' %}{{'<|assistant|>
+' + message['content'] + '<|end|>
+'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>
+' }}{% else %}{{ eos_token }}{% endif %}

config.json

@@ -0,0 +1,34 @@
+{
+  "architectures": [
+    "Phi3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoModelForCausalLM": "architecture.Phi3WithVectorMemoryForCausalLM"
+  },
+  "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,
+  "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": "4.52.4",
+  "use_cache": true,
+  "vocab_size": 32064
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 32000,
+  "pad_token_id": 32000,
+  "transformers_version": "4.52.4"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73c7d81226836583408bd6d5c9dc170c8190b92563d1d3d4dfd1ddbcb645ec31
+size 4977709096

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f311787aa136e858556caa8543015161edcad85ba81b6a36072443d7fa73c87
+size 2669692552

model.safetensors.index.json

@@ -0,0 +1,237 @@
+{
+  "metadata": {
+    "total_size": 7647373824
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.correction_head.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.correction_head.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.global_state_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.global_state_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.linear.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.local_state_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.local_state_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.decoder_attention.in_proj_bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.decoder_attention.in_proj_weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.decoder_attention.out_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.decoder_attention.out_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.decoder_ffn.0.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.decoder_ffn.0.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.decoder_ffn.2.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.decoder_ffn.2.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.decoder_layernorm.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.decoder_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.linear1.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.linear1.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.linear2.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.linear2.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.norm1.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.norm1.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.norm2.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.norm2.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.self_attn.in_proj_bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.self_attn.in_proj_weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.encoder.layers.0.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.memory_attention.in_proj_bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.memory_attention.in_proj_weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.memory_attention.out_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.memory_attention.out_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.memory_layernorm.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.memory_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.memory_queries": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,131 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "32000": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<|placeholder1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32003": {
+      "content": "<|placeholder2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32004": {
+      "content": "<|placeholder3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32005": {
+      "content": "<|placeholder4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32006": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32007": {
+      "content": "<|end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32008": {
+      "content": "<|placeholder5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32009": {
+      "content": "<|placeholder6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32010": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "legacy": false,
+  "model_max_length": 4096,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}