Upload fine-tuned model with ReflectiveMemoryLayer

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
+}

architectureV3.py

@@ -0,0 +1,218 @@
+# --- START OF FILE architectureV3.py ---
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from transformers import Phi3Config, Phi3ForCausalLM
+from transformers.modeling_outputs import CausalLMOutputWithPast
+from typing import Optional, Dict, Tuple
+from dataclasses import dataclass
+
+@dataclass
+class CausalLMOutputWithLTM(CausalLMOutputWithPast):
+    loss: Optional[torch.FloatTensor] = None
+    logits: torch.FloatTensor = None
+    past_key_values: Optional[Tuple[Tuple[torch.FloatTensor]]] = None
+    hidden_states: Optional[Tuple[torch.FloatTensor]] = None
+    attentions: Optional[Tuple[torch.FloatTensor]] = None
+    ltm_state: Optional[torch.Tensor] = None # The returned LTM state
+
+# --- BUILDING BLOCK 1: Hierarchical VectorMemoryHead (Stateless) ---
+class VectorMemoryHead(nn.Module):
+    def __init__(self, hidden_dim: int, num_memory_slots: int, num_heads: int, ff_dim: int,
+                 num_long_term_memory_slots: int = 0,
+                 device=None, dtype=None):
+        super().__init__()
+        self.hidden_dim = hidden_dim
+        self.num_memory_slots = num_memory_slots
+        self.num_long_term_memory_slots = num_long_term_memory_slots
+        self.use_long_term_memory = self.num_long_term_memory_slots > 0
+
+        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))
+
+        if self.use_long_term_memory:
+            self.memory_update_gate = nn.Sequential(
+                nn.Linear(hidden_dim, hidden_dim, device=device, dtype=dtype), nn.Sigmoid())
+            self.ltm_retrieval_attention = nn.MultiheadAttention(
+                embed_dim=hidden_dim, num_heads=num_heads, dropout=0.1, batch_first=True, device=device, dtype=dtype)
+
+    def forward(self, memory_input_sequence: torch.Tensor,
+                long_term_memory: Optional[torch.Tensor] = None) -> Tuple[torch.Tensor, torch.Tensor, Optional[torch.Tensor]]:
+        batch_size = memory_input_sequence.shape[0]
+        new_ltm_state = long_term_memory
+        queries = self.memory_queries.expand(batch_size, -1, -1)
+        encoded_vectors = self.encoder(memory_input_sequence)
+        compressed_memory, _ = self.memory_attention(query=queries, key=encoded_vectors, value=encoded_vectors)
+        compressed_memory = self.memory_layernorm(compressed_memory + queries)
+        final_memory_context = compressed_memory
+
+        if self.use_long_term_memory and long_term_memory is not None:
+            retrieved_ltm, _ = self.ltm_retrieval_attention(
+                query=compressed_memory, key=long_term_memory, value=long_term_memory)
+            l1_summary = compressed_memory.mean(dim=1, keepdim=True)
+            update_gate = self.memory_update_gate(l1_summary)
+            new_ltm_state = (update_gate * l1_summary) + ((1 - update_gate) * long_term_memory)
+            final_memory_context = final_memory_context + retrieved_ltm
+
+        reconstructed, _ = self.decoder_attention(query=encoded_vectors, key=final_memory_context, value=final_memory_context)
+        reconstructed_vectors = self.decoder_layernorm(reconstructed + encoded_vectors)
+        reconstructed_vectors = self.decoder_ffn(reconstructed_vectors)
+        return compressed_memory, reconstructed_vectors, new_ltm_state
+
+# --- BUILDING BLOCK 2: ReflectiveMemoryLayer ---
+class ReflectiveMemoryLayer(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, self.output_dim = original_layer.in_features, original_layer.out_features
+        self.memory_dim, self.global_state_storage = memory_dim, global_state_storage
+        self.linear = original_layer # Keep the original linear layer frozen
+        self.refinement_passes: int = 2
+        device, dtype = self.linear.weight.device, self.linear.weight.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, num_long_term_memory_slots=32, device=device, dtype=dtype)
+        self.thought_critique_attention = nn.MultiheadAttention(
+            embed_dim=memory_dim, num_heads=memory_num_heads, dropout=0.1, batch_first=True, device=device, dtype=dtype)
+        self.thought_layernorm = nn.LayerNorm(memory_dim, device=device, dtype=dtype)
+        self.correction_head = nn.Linear(memory_dim, 2 * self.output_dim, device=device, dtype=dtype)
+
+        self.last_corrected_activation, self.last_additive_correction = None, None
+        self.last_memory_input, self.last_reconstructed_from_memory = None, None
+
+    def forward(self, x: torch.Tensor):
+        base_output = self.linear(x)
+        if 'embeds' not in self.global_state_storage: 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
+        
+        ltm_state = self.global_state_storage.get('ltm', None)
+        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)
+
+        # *** FIX: Expand LTM state to match the flattened token dimension (B*S) ***
+        ltm_state_expanded = None
+        if ltm_state is not None:
+            ltm_state_expanded = ltm_state.repeat_interleave(S, dim=0)
+
+        compressed_mem_flat, recon_flat, new_ltm_state_expanded = self.memory_head(memory_input_flat, ltm_state_expanded)
+        
+        # *** FIX: Condense updated LTM state back to batch dimension B ***
+        if new_ltm_state_expanded is not None:
+             num_ltm_slots = new_ltm_state_expanded.shape[1]
+             new_ltm_condensed = new_ltm_state_expanded.view(B, S, num_ltm_slots, self.memory_dim).mean(dim=1)
+             self.global_state_storage['ltm'] = new_ltm_condensed.detach()
+
+        initial_thought = compressed_mem_flat.mean(dim=1).view(B, S, self.memory_dim)
+        current_thought = initial_thought
+        if not self.training and self.refinement_passes > 0:
+            with torch.no_grad():
+                for _ in range(self.refinement_passes):
+                    current_thought_flat = current_thought.view(B * S, 1, self.memory_dim)
+                    internal_ref, _ = self.memory_head.decoder_attention(
+                        query=current_thought_flat, key=compressed_mem_flat, value=compressed_mem_flat)
+                    external_crit, _ = self.thought_critique_attention(
+                        query=current_thought_flat, key=memory_input_flat, value=memory_input_flat)
+                    refined_thought = current_thought + internal_ref.view(B,S,-1) + external_crit.view(B,S,-1)
+                    current_thought = self.thought_layernorm(refined_thought)
+        
+        thought_for_correction = current_thought if not self.training else initial_thought
+        raw_correction = self.correction_head(thought_for_correction)
+        gate, value = torch.chunk(raw_correction, 2, dim=-1)
+        final_activation = base_output * torch.sigmoid(gate.to(x.dtype)) + value.to(x.dtype)
+        
+        if self.training:
+            self.last_corrected_activation = final_activation
+            self.last_additive_correction = value
+            self.last_memory_input = memory_input
+            self.last_reconstructed_from_memory = recon_flat.view(B, S, 2, self.memory_dim)
+        return final_activation
+
+# --- BUILDING BLOCK 3: The Full Custom Model with State Management ---
+class Phi3WithReflectiveMemoryForCausalLM(Phi3ForCausalLM):
+    def __init__(self, config):
+        super().__init__(config)
+        self.global_state_storage = {}
+        self.target_layer_path = "model.layers.15.mlp.gate_up_proj"
+        self.memory_dim, self.num_long_term_memory_slots = 64, 32
+
+        self.model.embed_tokens.register_forward_hook(
+            lambda module, input, output: self.global_state_storage.update({'embeds': output}))
+
+        try:
+            original_layer = self.get_submodule(self.target_layer_path)
+            custom_layer = ReflectiveMemoryLayer(
+                original_layer=original_layer, global_input_dim=config.hidden_size,
+                memory_dim=self.memory_dim, num_memory_slots=8, memory_num_heads=4,
+                global_state_storage=self.global_state_storage)
+            parent_path = ".".join(self.target_layer_path.split('.')[:-1])
+            setattr(self.get_submodule(parent_path), self.target_layer_path.split('.')[-1], custom_layer)
+            print(f"Successfully replaced '{self.target_layer_path}' with ReflectiveMemoryLayer.")
+        except AttributeError:
+            print(f"Could not find target layer '{self.target_layer_path}'. Model remains unmodified.")
+
+    def _init_ltm_state(self, batch_size, device, dtype):
+        # *** FIX: Initialize LTM state per item in the batch (no hardcoded hack) ***
+        return torch.zeros(
+            batch_size, self.num_long_term_memory_slots, self.memory_dim, device=device, dtype=dtype)
+
+    def forward(self, input_ids: torch.LongTensor = None, attention_mask: Optional[torch.Tensor] = None,
+                position_ids: Optional[torch.LongTensor] = None, past_key_values: Optional[list[torch.FloatTensor]] = None,
+                inputs_embeds: Optional[torch.FloatTensor] = None, labels: Optional[torch.LongTensor] = None,
+                use_cache: Optional[bool] = None, output_attentions: Optional[bool] = None,
+                output_hidden_states: Optional[bool] = None, return_dict: Optional[bool] = None,
+                ltm_state: Optional[torch.Tensor] = None):
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+        self.global_state_storage.clear()
+        
+        # *** FIX: Initialize LTM state if not provided, for both training and first step of inference ***
+        if ltm_state is None:
+             batch_size = input_ids.shape[0] if input_ids is not None else inputs_embeds.shape[0]
+             ltm_state = self._init_ltm_state(batch_size, self.device, self.dtype)
+        self.global_state_storage['ltm'] = ltm_state
+
+        outputs = self.model(
+            input_ids=input_ids, attention_mask=attention_mask, position_ids=position_ids,
+            past_key_values=past_key_values, inputs_embeds=inputs_embeds, use_cache=use_cache,
+            output_attentions=output_attentions, output_hidden_states=output_hidden_states, return_dict=return_dict)
+        
+        hidden_states = outputs[0]
+        logits = self.lm_head(hidden_states).float()
+
+        loss = None
+        if labels is not None:
+            loss_fct = nn.CrossEntropyLoss()
+            loss = loss_fct(logits[..., :-1, :].contiguous().view(-1, self.config.vocab_size),
+                            labels[..., 1:].contiguous().view(-1))
+            # Note: Auxiliary losses from main.py are calculated outside the model forward pass.
+
+        new_ltm_state = self.global_state_storage.get('ltm', None)
+        
+        if not return_dict:
+            output = (logits,) + outputs[1:] + (new_ltm_state,)
+            return (loss,) + output if loss is not None else output
+
+        return CausalLMOutputWithLTM(
+            loss=loss, logits=logits, past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states, attentions=outputs.attentions, ltm_state=new_ltm_state)

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 +1,35 @@
+{
+  "architectures": [
+    "Phi3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_phi3.Phi3Config",
+    "AutoModelForCausalLM": "architectureV3.Phi3WithReflectiveMemoryForCausalLM"
+  },
+  "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": "4.53.0",
+  "use_cache": true,
+  "vocab_size": 32064
+}

configuration_phi3.py

@@ -0,0 +1,227 @@
+# coding=utf-8
+# Copyright 2024 Microsoft and the HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+""" Phi-3 model configuration"""
+
+
+from transformers.configuration_utils import PretrainedConfig
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+PHI3_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+    "microsoft/Phi-3-mini-4k-instruct": "https://huggingface.co/microsoft/Phi-3-mini-4k-instruct/resolve/main/config.json",
+    "microsoft/Phi-3-mini-128k-instruct": "https://huggingface.co/microsoft/Phi-3-mini-128k-instruct/resolve/main/config.json",
+}
+
+
+class Phi3Config(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`Phi3Model`]. It is used to instantiate a Phi-3
+    model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
+    defaults will yield a similar configuration to that of the
+    [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct).
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+
+    Args:
+        vocab_size (`int`, *optional*, defaults to 32064):
+            Vocabulary size of the Phi-3 model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`Phi3Model`].
+        hidden_size (`int`, *optional*, defaults to 3072):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 8192):
+            Dimension of the MLP representations.
+        num_hidden_layers (`int`, *optional*, defaults to 32):
+            Number of hidden layers in the Transformer decoder.
+        num_attention_heads (`int`, *optional*, defaults to 32):
+            Number of attention heads for each attention layer in the Transformer decoder.
+        num_key_value_heads (`int`, *optional*):
+            This is the number of key_value heads that should be used to implement Grouped Query Attention. If
+            `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
+            `num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When
+            converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
+            by meanpooling all the original heads within that group. For more details checkout [this
+            paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
+            `num_attention_heads`.
+        resid_pdrop (`float`, *optional*, defaults to 0.0):
+            Dropout probability for mlp outputs.
+        embd_pdrop (`int`, *optional*, defaults to 0.0):
+            The dropout ratio for the embeddings.
+        attention_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio after computing the attention scores.
+        hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
+            The non-linear activation function (function or string) in the decoder.
+        max_position_embeddings (`int`, *optional*, defaults to 4096):
+            The maximum sequence length that this model might ever be used with.
+        original_max_position_embeddings (`int`, *optional*, defaults to 4096):
+            The maximum sequence length that this model was trained with. This is used to determine the size of the
+            original RoPE embeddings when using long scaling.
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        rms_norm_eps (`float`, *optional*, defaults to 1e-05):
+            The epsilon value used for the RMSNorm.
+        use_cache (`bool`, *optional*, defaults to `True`):
+            Whether or not the model should return the last key/values attentions (not used by all models). Only
+            relevant if `config.is_decoder=True`. Whether to tie weight embeddings or not.
+        tie_word_embeddings (`bool`, *optional*, defaults to `False`):
+            Whether to tie weight embeddings
+        rope_theta (`float`, *optional*, defaults to 10000.0):
+            The base period of the RoPE embeddings.
+        rope_scaling (`dict`, *optional*):
+            The scaling strategy for the RoPE embeddings. If `None`, no scaling is applied. If a dictionary, it must
+            contain the following keys: `type`, `short_factor` and `long_factor`. The `type` must be `longrope` and 
+            the `short_factor` and `long_factor` must be lists of numbers with the same length as the hidden size 
+            divided by the number of attention heads divided by 2.
+        bos_token_id (`int`, *optional*, defaults to 1):
+            The id of the "beginning-of-sequence" token.
+        eos_token_id (`int`, *optional*, defaults to 32000):
+            The id of the "end-of-sequence" token.
+        pad_token_id (`int`, *optional*, defaults to 32000):
+            The id of the padding token.
+        sliding_window (`int`, *optional*):
+            Sliding window attention window size. If `None`, no sliding window is applied.
+
+    Example:
+
+    ```python
+    >>> from transformers import Phi3Model, Phi3Config
+
+    >>> # Initializing a Phi-3 style configuration
+    >>> configuration = Phi3Config.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+
+    >>> # Initializing a model from the configuration
+    >>> model = Phi3Model(configuration)
+
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```"""
+
+    model_type = "phi3"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    def __init__(
+        self,
+        vocab_size=32064,
+        hidden_size=3072,
+        intermediate_size=8192,
+        num_hidden_layers=32,
+        num_attention_heads=32,
+        num_key_value_heads=None,
+        resid_pdrop=0.0,
+        embd_pdrop=0.0,
+        attention_dropout=0.0,
+        hidden_act="silu",
+        max_position_embeddings=4096,
+        original_max_position_embeddings=4096,
+        initializer_range=0.02,
+        rms_norm_eps=1e-5,
+        use_cache=True,
+        tie_word_embeddings=False,
+        rope_theta=10000.0,
+        rope_scaling=None,
+        bos_token_id=1,
+        eos_token_id=32000,
+        pad_token_id=32000,
+        sliding_window=None,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+
+        if num_key_value_heads is None:
+            num_key_value_heads = num_attention_heads
+
+        self.num_key_value_heads = num_key_value_heads
+        self.resid_pdrop = resid_pdrop
+        self.embd_pdrop = embd_pdrop
+        self.attention_dropout = attention_dropout
+        self.hidden_act = hidden_act
+        self.max_position_embeddings = max_position_embeddings
+        self.original_max_position_embeddings = original_max_position_embeddings
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = rms_norm_eps
+        self.use_cache = use_cache
+        self.rope_theta = rope_theta
+        self.rope_scaling = rope_scaling
+        self._rope_scaling_adjustment()
+        self._rope_scaling_validation()
+        self.sliding_window = sliding_window
+
+        super().__init__(
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            pad_token_id=pad_token_id,
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )
+
+    def _rope_scaling_adjustment(self):
+        """
+        Adjust the `type` of the `rope_scaling` configuration for backward compatibility.
+        """
+        if self.rope_scaling is None:
+            return
+
+        rope_scaling_type = self.rope_scaling.get("type", None)
+
+        # For backward compatibility if previous version used "su" or "yarn"
+        if rope_scaling_type is not None and rope_scaling_type in ["su", "yarn"]:
+            self.rope_scaling["type"] = "longrope"
+
+    def _rope_scaling_validation(self):
+        """
+        Validate the `rope_scaling` configuration.
+        """
+        if self.rope_scaling is None:
+            return
+
+        if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 3:
+            raise ValueError(
+                "`rope_scaling` must be a dictionary with three fields, `type`, `short_factor` and `long_factor`, "
+                f"got {self.rope_scaling}"
+            )
+        rope_scaling_type = self.rope_scaling.get("type", None)
+        rope_scaling_short_factor = self.rope_scaling.get("short_factor", None)
+        rope_scaling_long_factor = self.rope_scaling.get("long_factor", None)
+        if rope_scaling_type is None or rope_scaling_type not in ["longrope"]:
+            raise ValueError(f"`rope_scaling`'s type field must be one of ['longrope'], got {rope_scaling_type}")
+        if not (
+            isinstance(rope_scaling_short_factor, list)
+            and all(isinstance(x, (int, float)) for x in rope_scaling_short_factor)
+        ):
+            raise ValueError(
+                f"`rope_scaling`'s short_factor field must be a list of numbers, got {rope_scaling_short_factor}"
+            )
+        if not len(rope_scaling_short_factor) == self.hidden_size // self.num_attention_heads // 2:
+            raise ValueError(
+                f"`rope_scaling`'s short_factor field must have length {self.hidden_size // self.num_attention_heads // 2}, got {len(rope_scaling_short_factor)}"
+            )
+        if not (
+            isinstance(rope_scaling_long_factor, list)
+            and all(isinstance(x, (int, float)) for x in rope_scaling_long_factor)
+        ):
+            raise ValueError(
+                f"`rope_scaling`'s long_factor field must be a list of numbers, got {rope_scaling_long_factor}"
+            )
+        if not len(rope_scaling_long_factor) == self.hidden_size // self.num_attention_heads // 2:
+            raise ValueError(
+                f"`rope_scaling`'s long_factor field must have length {self.hidden_size // self.num_attention_heads // 2}, got {len(rope_scaling_long_factor)}"
+            )

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": "4.53.0"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b838acd9f2c8bbe02dedecef4234aeedf3304fbe7a1a786163300be81581def4
+size 4983487544

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 +1,250 @@
+{
+  "metadata": {
+    "total_parameters": 3826575232,
+    "total_size": 7653150464
+  },
+  "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.ltm_retrieval_attention.in_proj_bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.ltm_retrieval_attention.in_proj_weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.ltm_retrieval_attention.out_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.ltm_retrieval_attention.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.mlp.gate_up_proj.memory_head.memory_update_gate.0.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.memory_head.memory_update_gate.0.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.thought_critique_attention.in_proj_bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.thought_critique_attention.in_proj_weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.thought_critique_attention.out_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.thought_critique_attention.out_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.thought_layernorm.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.thought_layernorm.weight": "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
+}