Add files using upload-large-folder tool

README.md

@@ -0,0 +1,254 @@
+---
+license: mit
+library_name: transformers
+base_model:
+- deepseek-ai/DeepSeek-V3-Base
+---
+
+<p align="center">
+  <img src="images/deep-cogito-logo.png" alt="Logo" width="40%">
+</p>
+
+
+# Cogito v2 preview - 671B MoE (FP8)
+
+[Blog Post](https://www.deepcogito.com/research/cogito-v2-preview)
+
+The Cogito v2 LLMs are instruction tuned generative models. All models are released under an open license for commercial use.
+
+- Cogito v2 models are hybrid reasoning models. Each model can answer directly (standard LLM), or self-reflect before answering (like reasoning models).
+- The LLMs are trained using **Iterated Distillation and Amplification (IDA)** - an scalable and efficient alignment strategy for superintelligence using iterative self-improvement.
+- The models have been optimized for coding, STEM, instruction following and general helpfulness, and have significantly higher multilingual, coding and tool calling capabilities than size equivalent counterparts.
+  - In both standard and reasoning modes, Cogito v2-preview models outperform their size equivalent counterparts on common industry benchmarks. 
+- This model is trained in over 30 languages and supports a context length of 128k.
+
+# Evaluations
+For detailed evaluations, please refer to the [Blog Post](https://www.deepcogito.com/research/cogito-v2-preview).
+
+
+# Usage
+Here is a snippet below for usage with Transformers:
+
+```python
+import transformers
+import torch
+
+model_id = "deepcogito/cogito-v2-preview-llama-671B-MoE-FP8"
+
+pipeline = transformers.pipeline(
+    "text-generation",
+    model=model_id,
+    model_kwargs={"torch_dtype": torch.bfloat16},
+    device_map="auto",
+)
+
+messages = [
+    {"role": "system", "content": "You are a pirate chatbot who always responds in pirate speak!"},
+    {"role": "user", "content": "Give me a short introduction to LLMs."},
+]
+
+outputs = pipeline(
+    messages,
+    max_new_tokens=512,
+)
+
+print(outputs[0]["generated_text"][-1])
+```
+
+
+
+## Implementing extended thinking
+- By default, the model will answer in the standard mode. 
+- To enable thinking, you can do any one of the two methods:
+  - Set `enable_thinking=True` while applying the chat template.
+  - Add a specific system prompt, along with prefilling the response with "\<think\>\n". 
+
+**NOTE: Unlike Cogito v1 models, we initiate the response with "\<think\>\n" at the beginning of every output when reasoning is enabled. This is because hybrid models can be brittle at times (<0.1% of the cases), and adding a "\<think\>\n" ensures that the model does indeed respect thinking.**
+
+### Method 1 - Set enable_thinking=True in the tokenizer
+If you are using Huggingface tokenizers, then you can simply use add the argument `enable_thinking=True` to the tokenization (this option is added to the chat template).
+
+Here is an example - 
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model_name = "deepcogito/cogito-v2-preview-llama-671B-MoE-FP8"
+
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype="auto",
+    device_map="auto"
+)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+prompt = "Give me a short introduction to LLMs."
+messages = [
+    {"role": "system", "content": "You are a pirate chatbot who always responds in pirate speak!"},
+    {"role": "user", "content": prompt}
+]
+
+text = tokenizer.apply_chat_template(
+    messages,
+    tokenize=False,
+    add_generation_prompt=True,
+    enable_thinking=True
+)
+model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
+
+generated_ids = model.generate(
+    **model_inputs,
+    max_new_tokens=512
+)
+generated_ids = [
+    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
+]
+
+response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
+print(response)
+```
+
+### Method 2 - Add a specific system prompt, along with prefilling the response with "\<think\>\n". 
+To enable thinking using this method, you need to do two parts - 
+
+
+Step 1 - Simply use this in the system prompt `system_instruction = 'Enable deep thinking subroutine.'`
+
+If you already have a system_instruction, then use `system_instruction = 'Enable deep thinking subroutine.' + '\n\n' + system_instruction`.
+
+Step 2 - Prefil the response with the tokens `"<think>\n"`.
+
+Here is an example - 
+
+```python
+import transformers
+import torch
+
+model_name = "deepcogito/cogito-v2-preview-llama-109B-MoE"
+
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype="auto",
+    device_map="auto"
+)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+# Step 1 - Add deep thinking instruction.
+DEEP_THINKING_INSTRUCTION = "Enable deep thinking subroutine."
+
+messages = [
+    {"role": "system", "content": DEEP_THINKING_INSTRUCTION},
+    {"role": "user", "content": "Write a bash script that takes a matrix represented as a string with format '[1,2],[3,4],[5,6]' and prints the transpose in the same format."},
+]
+
+text = tokenizer.apply_chat_template(
+    messages,
+    tokenize=False,
+    add_generation_prompt=True
+)
+
+# Step 2 - Prefill response with "<think>\n".
+text += "<think>\n"
+
+# Now, continue as usual.
+model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
+
+generated_ids = model.generate(
+    **model_inputs,
+    max_new_tokens=512
+)
+generated_ids = [
+    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
+]
+
+response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
+print(response)
+```
+
+
+Similarly, if you have a system prompt, you can append the `DEEP_THINKING_INSTRUCTION` to the beginning in this way - 
+
+```python
+DEEP_THINKING_INSTRUCTION = "Enable deep thinking subroutine."
+
+system_prompt = "Reply to each prompt with only the actual code - no explanations."
+prompt = "Write a bash script that takes a matrix represented as a string with format '[1,2],[3,4],[5,6]' and prints the transpose in the same format."
+
+messages = [
+    {"role": "system", "content": DEEP_THINKING_INSTRUCTION + '\n\n' + system_prompt},
+    {"role": "user", "content": prompt}
+]
+```
+
+
+# Tool Calling
+Cogito models support tool calling (single, parallel, multiple and parallel_multiple) both in standard and extended thinking mode.
+
+Here is a snippet -
+
+```python
+# First, define a tool
+def get_current_temperature(location: str) -> float:
+    """
+    Get the current temperature at a location.
+    
+    Args:
+        location: The location to get the temperature for, in the format "City, Country"
+    Returns:
+        The current temperature at the specified location in the specified units, as a float.
+    """
+    return 22.  # A real function should probably actually get the temperature!
+
+# Next, create a chat and apply the chat template
+messages = [
+  {"role": "user", "content": "Hey, what's the temperature in Paris right now?"}
+]
+
+model_inputs = tokenizer.apply_chat_template(messages, tools=[get_current_temperature], add_generation_prompt=True)
+
+text = tokenizer.apply_chat_template(messages, tools=[get_current_temperature], add_generation_prompt=True, tokenize=False)
+inputs = tokenizer(text, return_tensors="pt", add_special_tokens=False).to(model.device)
+outputs = model.generate(**inputs, max_new_tokens=512)
+output_text = tokenizer.batch_decode(outputs)[0][len(text):]
+print(output_text)
+```
+
+This will result in the output - 
+```
+<｜tool▁calls▁begin｜><｜tool▁call▁begin｜>function<｜tool▁sep｜>get_current_temperature
+```json
+{"location":"Paris, France"}
+```<｜tool▁call▁end｜><｜tool▁calls▁end｜><｜end▁of▁sentence｜>
+```
+
+You can then generate text from this input as normal. If the model generates a tool call, you should add it to the chat like so:
+
+```python
+tool_call = {"name": "get_current_temperature", "arguments": {"location": "Paris, France"}}
+messages.append({"role": "assistant", "tool_calls": [{"type": "function", "function": tool_call}]})
+```
+
+and then call the tool and append the result, with the `tool` role, like so:
+
+```python
+messages.append({"role": "tool", "name": "get_current_temperature", "content": "22.0"})
+```
+
+After that, you can `generate()` again to let the model use the tool result in the chat:
+
+```python
+text = tokenizer.apply_chat_template(messages, tools=[get_current_temperature], add_generation_prompt=True, tokenize=False)
+inputs = tokenizer(text, return_tensors="pt", add_special_tokens=False).to(model.device)
+outputs = model.generate(**inputs, max_new_tokens=512)
+output_text = tokenizer.batch_decode(outputs)[0][len(text):]
+```
+
+This should result in the string -
+```
+'The current temperature in Paris is 22.0 degrees.<｜end▁of▁sentence｜>'
+```
+
+## License
+This repository and the model weights are licensed under **MIT License**.
+
+## Contact
+If you would like to reach out to our team, send an email to [contact@deepcogito.com](contact@deepcogito.com).

chat_template.jinja

@@ -0,0 +1,111 @@
+{# ==================================================================== #}
+{#  Deepseek v3 template with enable_thinking and tools support          #}
+{# ==================================================================== #}
+{%- if not enable_thinking is defined %}{% set enable_thinking = false %}{% endif -%}
+{%- if not tools is defined %}{% set tools = none %}{% endif -%}
+{%- if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif -%}
+
+{# --------------------------- Collect system prompt -------------------- #}
+{%- set ns = namespace(system_prompt='', is_last_user=false, outputs_open=false, first_output=true) -%}
+
+{%- if messages and messages[0].role == 'system' -%}
+  {%- set raw = messages[0].content -%}
+  {%- set ns.system_prompt = raw if raw is string else raw[0].text -%}
+  {%- set messages = messages[1:] -%}
+{%- endif -%}
+
+{# --------------------------- Inject deep thinking --------------------- #}
+{%- if enable_thinking -%}
+  {%- set ns.system_prompt = ns.system_prompt and 'Enable deep thinking subroutine.
+
+' ~ ns.system_prompt or 'Enable deep thinking subroutine.' -%}
+{%- endif -%}
+
+{# --------------------------- Append tools block ----------------------- #}
+{%- if tools is not none -%}
+  {%- if ns.system_prompt -%}
+    {%- set ns.system_prompt = ns.system_prompt ~ '
+
+You have the following functions available:
+
+' -%}
+  {%- else -%}
+    {%- set ns.system_prompt = 'You have the following functions available:
+
+' -%}
+  {%- endif -%}
+  {%- for t in tools -%}
+    {%- set ns.system_prompt = ns.system_prompt ~ "```json
+" ~ (t | tojson(indent=4)) ~ "
+```
+
+" -%}
+  {%- endfor -%}
+{%- endif -%}
+
+{{- bos_token -}}{{- ns.system_prompt -}}
+
+{# --------------------------- Iterate conversation --------------------- #}
+{%- for m in messages -%}
+  {# --------------------------- USER ---------------------------------- #}
+  {%- if m.role == 'user' -%}
+    {%- set ns.is_last_user = true -%}
+    {%- set txt = m.content if m.content is string else m.content | selectattr('type','equalto','text') | map(attribute='text') | join('') -%}
+    {{- "<｜User｜>" -}}{{- txt -}}{{- "<｜Assistant｜>" -}}
+  {%- endif -%}
+
+  {# --------------------------- ASSISTANT with TOOL CALLS -------------- #}
+  {%- if m.role == 'assistant' and m.tool_calls is defined and m.tool_calls -%}
+    {%- set ns.is_last_user = false -%}
+    {%- set lead = m.content is string and m.content|trim or (m.content and m.content | selectattr('type','equalto','text') | map(attribute='text') | join('')) or '' -%}
+    {{- lead -}}{{- "<｜tool▁calls▁begin｜>" -}}
+    {%- for call in m.tool_calls -%}
+      {{- "<｜tool▁call▁begin｜>" -}}{{- call.type -}}{{- "<｜tool▁sep｜>" -}}{{- call.function.name -}}
+      {{- "
+```json
+" -}}{{- call.function.arguments -}}{{- "
+```" -}}{{- "<｜tool▁call▁end｜>" -}}
+      {%- if not loop.last -%}{{- "
+" -}}{%- endif -%}
+    {%- endfor -%}
+    {{- "<｜tool▁calls▁end｜>" -}}{{- "<｜end▁of▁sentence｜>" -}}
+  {%- endif -%}
+
+  {# --------------------------- ASSISTANT plain ------------------------ #}
+  {%- if m.role == 'assistant' and (m.tool_calls is not defined or not m.tool_calls) -%}
+    {%- set ns.is_last_user = false -%}
+    {%- set txt = m.content if m.content is string else m.content | selectattr('type','equalto','text') | map(attribute='text') | join('') -%}
+    {{- txt -}}{{- "<｜end▁of▁sentence｜>" -}}
+  {%- endif -%}
+
+  {# --------------------------- TOOL output ---------------------------- #}
+  {%- if m.role == 'tool' -%}
+    {%- set ns.is_last_user = false -%}
+    {%- set out_txt = m.content if m.content is string else m.content | selectattr('type','equalto','text') | map(attribute='text') | join('') -%}
+    {%- if not ns.outputs_open -%}
+      {{- "<｜tool▁outputs▁begin｜>" -}}
+      {%- set ns.outputs_open = true -%}
+    {%- endif -%}
+    {{- "<｜tool▁output▁begin｜>" -}}{{- out_txt -}}{{- "<｜tool▁output▁end｜>" -}}
+    {%- if loop.nextitem is defined and loop.nextitem.role == 'tool' -%}
+      {{- "
+" -}}
+    {%- endif -%}
+    {%- if loop.nextitem is undefined or loop.nextitem.role != 'tool' -%}
+      {{- "<｜tool▁outputs▁end｜>" -}}
+      {%- set ns.outputs_open = false -%}
+    {%- endif -%}
+  {%- endif -%}
+{%- endfor -%}
+
+{%- if ns.outputs_open -%}
+  {{- "<｜tool▁outputs▁end｜>" -}}
+{%- endif -%}
+
+{%- if add_generation_prompt and not ns.is_last_user -%}
+  {{- "<｜Assistant｜>" -}}
+{%- endif -%}
+
+{%- if add_generation_prompt and enable_thinking -%}
+  {{- '<think>\n' -}}
+{%- endif -%}

config.json

@@ -0,0 +1,107 @@
+{
+  "architectures": [
+    "DeepseekV3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_deepseek.DeepseekV3Config",
+    "AutoModel": "modeling_deepseek.DeepseekV3Model",
+    "AutoModelForCausalLM": "modeling_deepseek.DeepseekV3ForCausalLM"
+  },
+  "aux_loss_alpha": 0.001,
+  "bos_token_id": 0,
+  "eos_token_id": 1,
+  "ep_size": 1,
+  "first_k_dense_replace": 3,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 7168,
+  "initializer_range": 0.02,
+  "intermediate_size": 18432,
+  "kv_lora_rank": 512,
+  "max_position_embeddings": 163840,
+  "model_type": "deepseek_v3",
+  "moe_intermediate_size": 2048,
+  "moe_layer_freq": 1,
+  "n_group": 8,
+  "n_routed_experts": 256,
+  "n_shared_experts": 1,
+  "norm_topk_prob": true,
+  "num_attention_heads": 128,
+  "num_experts_per_tok": 8,
+  "num_hidden_layers": 61,
+  "num_key_value_heads": 128,
+  "num_nextn_predict_layers": 1,
+  "pretraining_tp": 1,
+  "q_lora_rank": 1536,
+  "qk_head_dim": 192,
+  "qk_nope_head_dim": 128,
+  "qk_rope_head_dim": 64,
+  "quantization_config": {
+    "config_groups": {
+      "group_0": {
+        "input_activations": {
+          "actorder": null,
+          "block_structure": null,
+          "dynamic": true,
+          "group_size": null,
+          "num_bits": 8,
+          "observer": null,
+          "observer_kwargs": {},
+          "strategy": "token",
+          "symmetric": true,
+          "type": "float"
+        },
+        "output_activations": null,
+        "targets": [
+          "Linear"
+        ],
+        "weights": {
+          "actorder": null,
+          "block_structure": null,
+          "dynamic": false,
+          "group_size": null,
+          "num_bits": 8,
+          "observer": "minmax",
+          "observer_kwargs": {},
+          "strategy": "channel",
+          "symmetric": true,
+          "type": "float"
+        }
+      }
+    },
+    "format": "float-quantized",
+    "global_compression_ratio": null,
+    "ignore": [
+      "lm_head"
+    ],
+    "kv_cache_scheme": null,
+    "quant_method": "compressed-tensors",
+    "quantization_status": "compressed"
+  },
+  "rms_norm_eps": 1e-06,
+  "rope_interleave": true,
+  "rope_scaling": {
+    "beta_fast": 32.0,
+    "beta_slow": 1.0,
+    "factor": 40.0,
+    "mscale": 1.0,
+    "mscale_all_dim": 1.0,
+    "original_max_position_embeddings": 4096,
+    "rope_type": "yarn",
+    "type": "yarn"
+  },
+  "rope_theta": 10000,
+  "routed_scaling_factor": 2.5,
+  "scoring_func": "sigmoid",
+  "seq_aux": true,
+  "tie_word_embeddings": false,
+  "topk_group": 4,
+  "topk_method": "noaux_tc",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.53.0",
+  "use_cache": true,
+  "v_head_dim": 128,
+  "vocab_size": 128815
+}

configuration_deepseek.py

@@ -0,0 +1,210 @@
+from transformers.configuration_utils import PretrainedConfig
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+DEEPSEEK_PRETRAINED_CONFIG_ARCHIVE_MAP = {}
+class DeepseekV3Config(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`DeepseekV3Model`]. It is used to instantiate an DeepSeek
+    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 DeepSeek-V3.
+
+    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 129280):
+            Vocabulary size of the Deep model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`DeepseekV3Model`]
+        hidden_size (`int`, *optional*, defaults to 4096):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 11008):
+            Dimension of the MLP representations.
+        moe_intermediate_size (`int`, *optional*, defaults to 1407):
+            Dimension of the MoE representations.
+        num_hidden_layers (`int`, *optional*, defaults to 32):
+            Number of hidden layers in the Transformer decoder.
+        num_nextn_predict_layers (`int`, *optional*, defaults to 1):
+            Number of nextn predict layers in the DeepSeekV3 Model.
+        num_attention_heads (`int`, *optional*, defaults to 32):
+            Number of attention heads for each attention layer in the Transformer decoder.
+        n_shared_experts (`int`, *optional*, defaults to None):
+            Number of shared experts, None means dense model.
+        n_routed_experts (`int`, *optional*, defaults to None):
+            Number of routed experts, None means dense model.
+        routed_scaling_factor (`float`, *optional*, defaults to 1.0):
+            Scaling factor or routed experts.
+        topk_method (`str`, *optional*, defaults to `gready`):
+            Topk method used in routed gate.
+        n_group (`int`, *optional*, defaults to None):
+            Number of groups for routed experts.
+        topk_group (`int`, *optional*, defaults to None):
+            Number of selected groups for each token(for each token, ensuring the selected experts is only within `topk_group` groups).
+        num_experts_per_tok (`int`, *optional*, defaults to None):
+            Number of selected experts, None means dense model.
+        moe_layer_freq (`int`, *optional*, defaults to 1):
+            The frequency of the MoE layer: one expert layer for every `moe_layer_freq - 1` dense layers.
+        first_k_dense_replace (`int`, *optional*, defaults to 0):
+            Number of dense layers in shallow layers(embed->dense->dense->...->dense->moe->moe...->lm_head).
+                                                            \--k dense layers--/
+        norm_topk_prob (`bool`, *optional*, defaults to False):
+            Whether to normalize the weights of the routed experts.
+        scoring_func (`str`, *optional*, defaults to 'softmax'):
+            Method of computing expert weights.
+        aux_loss_alpha (`float`, *optional*, defaults to 0.001):
+            Auxiliary loss weight coefficient.
+        seq_aux = (`bool`, *optional*, defaults to True):
+            Whether to compute the auxiliary loss for each individual sample.
+        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`.
+        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 2048):
+            The maximum sequence length that this model might ever be used with.
+        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-06):
+            The epsilon used by the rms normalization layers.
+        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`.
+        pad_token_id (`int`, *optional*):
+            Padding token id.
+        bos_token_id (`int`, *optional*, defaults to 1):
+            Beginning of stream token id.
+        eos_token_id (`int`, *optional*, defaults to 2):
+            End of stream token id.
+        pretraining_tp (`int`, *optional*, defaults to 1):
+            Experimental feature. Tensor parallelism rank used during pretraining. Please refer to [this
+            document](https://huggingface.co/docs/transformers/parallelism) to understand more about it. This value is
+            necessary to ensure exact reproducibility of the pretraining results. Please refer to [this
+            issue](https://github.com/pytorch/pytorch/issues/76232).
+        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*):
+            Dictionary containing the scaling configuration for the RoPE embeddings. Currently supports two scaling
+            strategies: linear and dynamic. Their scaling factor must be a float greater than 1. The expected format is
+            `{"type": strategy name, "factor": scaling factor}`. When using this flag, don't update
+            `max_position_embeddings` to the expected new maximum.
+        attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`):
+            Whether to use a bias in the query, key, value and output projection layers during self-attention.
+        attention_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio for the attention probabilities.
+
+    ```python
+    >>> from transformers import DeepseekV3Model, DeepseekV3Config
+
+    >>> # Initializing a Deepseek-V3 style configuration
+    >>> configuration = DeepseekV3Config()
+
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```"""
+
+    model_type = "deepseek_v3"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    def __init__(
+        self,
+        vocab_size=129280,
+        hidden_size=7168,
+        intermediate_size=18432,
+        moe_intermediate_size = 2048,
+        num_hidden_layers=61,
+        num_nextn_predict_layers=1,
+        num_attention_heads=128,
+        num_key_value_heads=128,
+        n_shared_experts = 1,
+        n_routed_experts = 256,
+        ep_size = 1,
+        routed_scaling_factor = 2.5,
+        kv_lora_rank = 512,
+        q_lora_rank = 1536,
+        qk_rope_head_dim = 64,
+        v_head_dim = 128,
+        qk_nope_head_dim = 128,
+        topk_method = 'noaux_tc',
+        n_group = 8,
+        topk_group = 4,
+        num_experts_per_tok = 8,
+        moe_layer_freq = 1,
+        first_k_dense_replace = 3,
+        norm_topk_prob = True,
+        scoring_func = 'sigmoid',
+        aux_loss_alpha = 0.001,
+        seq_aux = True,
+        hidden_act="silu",
+        max_position_embeddings=4096,
+        initializer_range=0.02,
+        rms_norm_eps=1e-6,
+        use_cache=True,
+        pad_token_id=None,
+        bos_token_id=0,
+        eos_token_id=1,
+        pretraining_tp=1,
+        tie_word_embeddings=False,
+        rope_theta=10000.0,
+        rope_scaling=None,
+        attention_bias=False,
+        attention_dropout=0.0,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.max_position_embeddings = max_position_embeddings
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.moe_intermediate_size = moe_intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_nextn_predict_layers = num_nextn_predict_layers
+        self.num_attention_heads = num_attention_heads
+        self.n_shared_experts = n_shared_experts
+        self.n_routed_experts = n_routed_experts
+        self.ep_size = ep_size
+        self.routed_scaling_factor = routed_scaling_factor
+        self.kv_lora_rank = kv_lora_rank
+        self.q_lora_rank = q_lora_rank
+        self.qk_rope_head_dim = qk_rope_head_dim
+        self.v_head_dim = v_head_dim
+        self.qk_nope_head_dim = qk_nope_head_dim
+        self.topk_method = topk_method
+        self.n_group = n_group
+        self.topk_group = topk_group
+        self.num_experts_per_tok = num_experts_per_tok
+        self.moe_layer_freq = moe_layer_freq
+        self.first_k_dense_replace = first_k_dense_replace
+        self.norm_topk_prob = norm_topk_prob
+        self.scoring_func = scoring_func
+        self.aux_loss_alpha = aux_loss_alpha
+        self.seq_aux = seq_aux
+        # for backward compatibility
+        if num_key_value_heads is None:
+            num_key_value_heads = num_attention_heads
+
+        self.num_key_value_heads = num_key_value_heads
+        self.hidden_act = hidden_act
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = rms_norm_eps
+        self.pretraining_tp = pretraining_tp
+        self.use_cache = use_cache
+        self.rope_theta = rope_theta
+        self.rope_scaling = rope_scaling
+        self.attention_bias = attention_bias
+        self.attention_dropout = attention_dropout
+
+        super().__init__(
+            pad_token_id=pad_token_id,
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 1,
+  "transformers_version": "4.53.0"
+}

model-00001-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6d7bb74d1f02537f496cb82bdfd2056dd7ed89b10140c59a872a15e31d1792f
+size 4989505008

model-00002-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1f1cca90647254846a50de1215c43ae8381f3ef22e4b2e0741fd13e20a3b57e
+size 4993855904

model-00003-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d520e4dd1cadcf7bf4c29e61327581fc6cf614aff11adf96f09f1fd941b75ae8
+size 4993866592

model-00007-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:065af7fe792a792f847c45cd6aabfdf7b3dc113e8e3508a18c40666fe15bb8b6
+size 4993856176

model-00010-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc54c84d60133cb7a3c8aaa9a0386eed0bb4807614df2648eca43f2627124925
+size 4993861744

model-00011-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e98396d46e049b8774b7fddc3843cb5b06bc8b478235c2729f2f0c5ba8a95ff0
+size 4993855712

model-00012-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75c1a9f889cf3ef9fdcd3ed1faa5c16b28ad97def38c7c0959a7553e3235852a
+size 4993856352

model-00014-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd9fd84e4cf22ecfaa3f3160fb9a012c323cbb668ad3a8918695c23c4de45cdd
+size 4993866472

model-00016-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22d772e564b84999c6ebeac28ac0e7dc5f5ff61a2bad84f6c3a6f56ffc3c7c44
+size 4993856024

model-00031-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28360c7e425c31eea25c36998f4d08772108af978356cca0fbc3bc81ac2076df
+size 4993862192

model-00032-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5ffc41db9d08b624230d1b8c86479ce521c3f10a17a41b9029e08f99f8b304b
+size 4993856608

model-00040-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc7e584539a35601527f15ba76b340a78c2c9af174fd534b1bbf48ab1ba76529
+size 4993862376

model-00042-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:faa3df45d5f7ee90fec7da68050e3b59bf4f31ffdd05fbac6a1af714543d9120
+size 4993857032

model-00043-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc554ea5ec1e5dafaaed4e302841dee4ee26aa6689f69feadb81f6d38a54c158
+size 4993861984

model-00046-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef85409808b9c5a095b6d1119d8f8e87e26eed0aa71e5b019112c8cda907583b
+size 4993866960

model-00048-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20045452dfd6398a358ddc1fb54412fe3d2b4ada9ea05f3ee74d02629e5b6f26
+size 4993856512

model-00049-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15b07911e0fb80c01bdd09e1f93aae78865f52d5cefd4e3fed235557270f3464
+size 4993857032

model-00055-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb53d1bb43669ab85b19b14f43b27e64f0160b9f4f0f1256acdb8c29ad23d85c
+size 4993856560

model-00058-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb9edcb6e48608540eee80076f0f3120052521fd041842d3ab33847aa0636dfb
+size 4993857032

model-00060-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c05568c53c226911bcdbab0f0e5c8bc8d5fb9728a7461759495e614a6ced76bd
+size 4993856832

model-00063-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b459449fee2cb54b90c3ed4d1dd6c7811097d2eb95dbc2b603b27fb2987dbdcf
+size 4964534304

model-00064-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54921ff672a942a3e16c78caa24c6e69d3738a9e6e6e74966a532fde19201118
+size 4993805424

model-00071-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77962f002abbcdc4af6d554403f9396349673c9e21c7b6a58f3daaa28bf08f89
+size 4993866688

model-00072-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05fb152e3823e0421b256f8b1f703366574935810bfdc4de386bb13998459d6d
+size 4993857032

model-00079-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44eb49ee6bd4674b5b5b866124e6b4535792f7f535aed44857206f7930ad2817
+size 4993857032

model-00080-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e308c88f7b61ac6241d602e11a3b74cf43a3b577de886f88a87820c2c2a70a4a
+size 4993861984

model-00084-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a221b09968605bf9796e7889a662c0fc2674b1e4ee98a51cbe7f89fbc12a9c6
+size 4993862224

model-00090-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ce0afb7b5e419175d6a829c4ec69fa44e44b72b73f91d8e7f6c46604bf1f37d
+size 4993856840

model-00095-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ece1ccb5fa55e9910c9bec1ebb41d6a159784d511d837c67f6e1bc30734c663c
+size 4993857032

model-00099-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3654c3c895b8a25e8c2a45e2abac0973674bfae051902c44df52a56303d0b525
+size 4993856696

model-00107-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:171a04e26d70d3281369c5c1364b98647908066257a05a92a1242adb1cb98814
+size 4993862264

model-00111-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c8ff905d747f3ff91e6cd0c235ed12ed26c9d1fb20e9c3505ef019dda1acee7
+size 4993857008

model-00112-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:314e4165a976c71a7842d9522da3acf83bc92f0e37d52f255782eebe38b43116
+size 4993862000

model-00117-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7707ecbf184f8d14eb6ce2417f42527f071fcfada9ec5e0698ae4bab73b9354a
+size 4993862032

model-00126-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb82c6b60cc68b01b0181c7f4a2fa6a0ee0ebf2d73162a85800fac6e5dd0228f
+size 4993861984

model-00128-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14ba895c088d2334a75ef65a51b12e36fda82245e5078f5903273ab8970aeeb5
+size 4993862096

model-00130-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:166b10be78bcb8cb9c0a5b4157880e8c4688672ac13709551556b0eac9aea0d5
+size 4993862304

model-00132-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4e3c717253042a1d2ae43b850c1bb558bf4e7f6860581e33b90e4b894bf4ab4
+size 4993857032

model-00134-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef233e025de65e86d73ddc08d531d27a338724b19339d453ba538ace8e508004
+size 4993867216

model-00135-of-00135.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c1b8d17dc2cb1055088e76568eebd068cd57282297ec01c0d580564bb1f9aab
+size 4185771368

recipe.yaml

@@ -0,0 +1,6 @@
+default_stage:
+  default_modifiers:
+    QuantizationModifier:
+      targets: [Linear]
+      ignore: [lm_head]
+      scheme: FP8_DYNAMIC

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<｜begin▁of▁sentence｜>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<｜end▁of▁sentence｜>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<｜▁pad▁｜>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}