Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

chat_template.jinja

@@ -0,0 +1,209 @@
+{% macro render_extra_keys(json_dict, handled_keys) %}
+    {%- if json_dict is mapping %}
+        {%- for json_key in json_dict if json_key not in handled_keys %}
+            {%- if json_dict[json_key] is mapping or (json_dict[json_key] is sequence and json_dict[json_key] is not string) %}
+                {{- '\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | tojson | safe) ~ '</' ~ json_key ~ '>' }}
+            {%- else %}
+                {{-'\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | string) ~ '</' ~ json_key ~ '>' }}
+            {%- endif %}
+        {%- endfor %}
+    {%- endif %}
+{% endmacro %}
+{%- set enable_thinking = enable_thinking if enable_thinking is defined else True %}
+{%- set reasoning_budget = reasoning_budget if reasoning_budget is defined else None %}
+{%- set truncate_history_thinking = truncate_history_thinking if truncate_history_thinking is defined else True %}
+
+{%- set ns = namespace(last_user_idx = -1) %}
+{%- set loop_messages = messages %}
+{%- for m in loop_messages %}
+  {%- if m["role"] == "user" %}
+    {%- set ns.last_user_idx = loop.index0 %}
+  {%- endif %}
+{%- endfor %}
+
+{%- if messages[0]["role"] == "system" %}
+    {%- set system_message = messages[0]["content"] %}
+    {%- set loop_messages = messages[1:] %}
+{%- else %}
+    {%- set system_message = "" %}
+    {%- set loop_messages = messages %}
+{%- endif %}
+{%- if not tools is defined %}
+    {%- set tools = [] %}
+{%- endif %}
+{# Recompute last_user_idx relative to loop_messages after handling system #}
+{%- set ns = namespace(last_user_idx = -1) %}
+{%- for m in loop_messages %}
+  {%- if m["role"] == "user" %}
+    {%- set ns.last_user_idx = loop.index0 %}
+  {%- endif %}
+{%- endfor %}
+{%- if system_message is defined %}
+    {{- "<|im_start|>system\n" + system_message }}
+{%- else %}
+    {%- if tools is iterable and tools | length > 0 %}
+        {{- "<|im_start|>system\n" }}
+    {%- endif %}
+{%- endif %}
+{%- if tools is iterable and tools | length > 0 %}
+    {%- if system_message is defined and system_message | length > 0 %}
+        {{- "\n\n" }}
+    {%- endif %}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n" }}
+    {{- "<tools>" }}
+    {%- for tool in tools %}
+        {%- if tool.function is defined %}
+            {%- set tool = tool.function %}
+        {%- endif %}
+        {{- "\n<function>\n<name>" ~ tool.name ~ "</name>" }}
+        {%- if tool.description is defined %}
+            {{- '\n<description>' ~ (tool.description | trim) ~ '</description>' }}
+        {%- endif %}
+        {{- '\n<parameters>' }}
+        {%- if tool.parameters is defined and tool.parameters is mapping and tool.parameters.properties is defined and tool.parameters.properties is mapping %}
+            {%- for param_name, param_fields in tool.parameters.properties|items %}
+                {{- '\n<parameter>' }}
+                {{- '\n<name>' ~ param_name ~ '</name>' }}
+                {%- if param_fields.type is defined %}
+                    {{- '\n<type>' ~ (param_fields.type | string) ~ '</type>' }}
+                {%- endif %}
+                {%- if param_fields.description is defined %}
+                    {{- '\n<description>' ~ (param_fields.description | trim) ~ '</description>' }}
+                {%- endif %}
+                {%- if param_fields.enum is defined %}
+                    {{- '\n<enum>' ~ (param_fields.enum | tojson | safe) ~ '</enum>' }}
+                {%- endif %}
+                {%- set handled_keys = ['name', 'type', 'description', 'enum'] %}
+                {{- render_extra_keys(param_fields, handled_keys) }}
+                {{- '\n</parameter>' }}
+            {%- endfor %}
+        {%- endif %}
+        {% set handled_keys = ['type', 'properties', 'required'] %}
+        {{- render_extra_keys(tool.parameters, handled_keys) }}
+        {%- if tool.parameters is defined and tool.parameters.required is defined %}
+            {{- '\n<required>' ~ (tool.parameters.required | tojson | safe) ~ '</required>' }}
+        {%- endif %}
+        {{- '\n</parameters>' }}
+        {%- set handled_keys = ['type', 'name', 'description', 'parameters'] %}
+        {{- render_extra_keys(tool, handled_keys) }}
+        {{- '\n</function>' }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+
+    {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
+{%- endif %}
+
+
+{%- if system_message is defined %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if tools is iterable and tools | length > 0 %}
+        {{- '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+
+{%- for message in loop_messages %}
+    {%- if message.role == "assistant" %}
+        {# Add reasoning content in to content field for unified processing below. #}
+        {%- if message.reasoning_content is defined and message.reasoning_content is string and message.reasoning_content | trim | length > 0 %}
+            {%- set content = "<think>\n" ~ message.reasoning_content ~ "\n</think>\n" ~ (message.content | default('', true)) %}
+        {%- else %}
+            {%- set content = message.content | default('', true) %}
+            {%- if content is string -%}
+                {# Allow downstream logic to to take care of broken thought, only handle coherent reasoning here. #}
+                {%- if '<think>' not in content and '</think>' not in content -%}
+                    {%- set content = "<think></think>" ~ content -%}
+                {%- endif -%}
+            {%- else -%}
+                {%- set content = content -%}
+            {%- endif -%}
+        {%- endif %}
+        {%- if message.tool_calls is defined and message.tool_calls is iterable and message.tool_calls | length > 0 %}
+            {# Assistant message has tool calls. #}
+            {{- '<|im_start|>assistant\n' }}
+                {%- set include_content = not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}
+                {%- if content is string and content | trim | length > 0 %}
+                    {%- if include_content %}
+                        {{- (content | trim) ~ '\n' -}}
+                    {%- else %}
+                        {%- set c = (content | string) %}
+                        {%- if '</think>' in c %}
+                            {# Keep only content after the last closing think. Also generation prompt causes this. #}
+                            {%- set c = c.split('</think>')[-1] %}
+                        {%- elif '<think>' in c %}
+                            {# If <think> was opened but never closed, drop the trailing think segment #}
+                            {%- set c = c.split('<think>')[0] %}
+                        {%- endif %}
+                        {%- set c = "<think></think>" ~ c | trim %}
+                        {%- if c | length > 0 %}
+                            {{- c ~ '\n' -}}
+                        {%- endif %}
+                    {%- endif %}
+                {%- else %}
+                    {{- "<think></think>" -}}
+                {%- endif %}
+                {%- for tool_call in message.tool_calls %}
+                    {%- if tool_call.function is defined %}
+                        {%- set tool_call = tool_call.function %}
+                    {%- endif %}
+                    {{- '<tool_call>\n<function=' ~ tool_call.name ~ '>\n' -}}
+                        {%- if tool_call.arguments is defined %}
+                            {%- for args_name, args_value in tool_call.arguments|items %}
+                                {{- '<parameter=' ~ args_name ~ '>\n' -}}
+                                    {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
+                                {{- args_value ~ '\n</parameter>\n' -}}
+                            {%- endfor %}
+                        {%- endif %}
+                    {{- '</function>\n</tool_call>\n' -}}
+                {%- endfor %}
+                {{- '<|im_end|>\n' }}
+        {%- else %}
+            {# Assistant message doesn't have tool calls. #}
+            {%- if not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}
+                {{- '<|im_start|>assistant\n' ~ (content | default('', true) | string | trim) ~ '<|im_end|>\n' }}
+            {%- else %}
+                {%- set c = (content | default('', true) | string) %}
+                {%- if '<think>' in c and '</think>' in c %}
+                    {%- set c = "<think></think>" ~ c.split('</think>')[-1] %}
+                {%- endif %}
+                {%- set c = c | trim %}
+                {%- if c | length > 0 %}
+                    {{- '<|im_start|>assistant\n' ~ c ~ '<|im_end|>\n' }}
+                {%- else %}
+                    {{- '<|im_start|>assistant\n<|im_end|>\n' }}
+                {%- endif %}
+            {%- endif %}
+        {%- endif %}
+    {%- elif message.role == "user" or message.role == "system" %}
+        {{- '<|im_start|>' + message.role + '\n' }}
+        {%- set content = message.content | string %}
+        {%- if message.role == "user" and loop.index0 == ns.last_user_idx and reasoning_budget is not none %}
+            {{- content + '\n\n{thinking token budget: ' + (reasoning_budget | string) + '}' }}
+        {%- else %}
+            {{- content }}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user\n' }}
+        {%- endif %}
+        {{- '<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>\n' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endfor %}
+
+{%- if add_generation_prompt %}
+    {%- if enable_thinking %}
+        {{- '<|im_start|>assistant\n<think>\n' }}
+    {%- else %}
+        {{- '<|im_start|>assistant\n<think></think>' }}
+    {%- endif %}
+{%- endif %}

config.json

@@ -0,0 +1,70 @@
+{
+  "architectures": [
+    "NemotronHForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_nemotron_h.NemotronHConfig",
+    "AutoModel": "modeling_nemotron_h.NemotronHForCausalLM",
+    "AutoModelForCausalLM": "modeling_nemotron_h.NemotronHForCausalLM"
+  },
+  "bos_token_id": 1,
+  "chunk_size": 128,
+  "conv_kernel": 4,
+  "eos_token_id": 2,
+  "expand": 2,
+  "head_dim": 128,
+  "hidden_dropout": 0.0,
+  "hidden_size": 2688,
+  "hybrid_override_pattern": "MEMEM*EMEMEM*EMEMEM*EMEMEM*EMEMEM*EMEMEMEM*EMEMEMEME",
+  "initializer_range": 0.02,
+  "intermediate_size": 1856,
+  "layer_norm_epsilon": 1e-05,
+  "mamba_head_dim": 64,
+  "mamba_hidden_act": "silu",
+  "mamba_num_heads": 64,
+  "mamba_proj_bias": false,
+  "mamba_ssm_cache_dtype": "float32",
+  "max_position_embeddings": 262144,
+  "mlp_bias": false,
+  "mlp_hidden_act": "relu2",
+  "model_type": "nemotron_h",
+  "moe_intermediate_size": 1856,
+  "moe_shared_expert_intermediate_size": 3712,
+  "n_group": 1,
+  "n_groups": 8,
+  "n_routed_experts": 128,
+  "n_shared_experts": 1,
+  "norm_eps": 1e-05,
+  "norm_topk_prob": true,
+  "num_attention_heads": 32,
+  "num_experts_per_tok": 6,
+  "num_hidden_layers": 52,
+  "num_key_value_heads": 2,
+  "num_logits_to_keep": 1,
+  "pad_token_id": 0,
+  "partial_rotary_factor": 1.0,
+  "rescale_prenorm_residual": true,
+  "residual_in_fp32": false,
+  "rope_theta": 10000,
+  "routed_scaling_factor": 2.5,
+  "sliding_window": null,
+  "ssm_state_size": 128,
+  "tie_word_embeddings": false,
+  "time_step_floor": 0.0001,
+  "time_step_limit": [
+    0.0,
+    Infinity
+  ],
+  "time_step_max": 0.1,
+  "time_step_min": 0.001,
+  "topk_group": 1,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.53.2",
+  "use_bias": false,
+  "use_cache": true,
+  "use_conv_bias": true,
+  "use_mamba_kernels": true,
+  "vocab_size": 131072
+}

configuration_nemotron_h.py

@@ -0,0 +1,262 @@
+# coding=utf-8
+# Copyright 2024 AI21 Labs Ltd. and the HuggingFace Inc. team. All rights reserved.
+# Copyright (c) 2025, NVIDIA CORPORATION. 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.
+"""NemotronH model configuration"""
+
+import re
+
+from transformers.configuration_utils import PretrainedConfig
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+
+class NemotronHConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`NemotronHModel`]. It is used to instantiate a
+    NemotronH 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 NemotronH-v0.1 model.
+
+    [todo](todo)
+
+    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 131072):
+            Vocabulary size of the NemotronH model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`NemotronHModel`]
+        tie_word_embeddings (`bool`, *optional*, defaults to `False`):
+            Whether the model's input and output word embeddings should be tied. Note that this is only relevant if the
+            model has a output word embedding layer.
+        hidden_size (`int`, *optional*, defaults to 4096):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 21504):
+            Dimension of the MLP representations.
+        num_hidden_layers (`int`, *optional*, defaults to 52):
+            Number of hidden layers in the Transformer encoder.
+        hybrid_override_pattern (`str`, *optional*, defaults to `"M-M-M-M*-M-M-M-M-M*-M-M-M-M-M*-M-M-M-M-M*-M-M-M-M-M-"`):
+            The pattern of the hybrid model. The pattern is a string of characters where each character represents M: Mamba2, *: Attention, -: MLP
+        num_attention_heads (`int`, *optional*, defaults to 32):
+            Number of attention heads for each attention layer in the Transformer encoder.
+        head_dim (`int`, *optional*, defaults to 128):
+            Dimension of each attention head.
+        num_key_value_heads (`int`, *optional*, defaults to 8):
+            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.
+        mlp_hidden_act (`str`, *optional*, defaults to "relu2"):
+            The non-linear activation function in the MLP layers.
+        attention_bias (`bool`, *optional*, defaults to `False`):
+            Whether to use bias in attention layers.
+        mlp_bias (`bool`, *optional*, defaults to `False`):
+            Whether to use bias in MLP layers.
+        use_bias (`bool`, *optional*, defaults to `False`):
+            Whether to use bias in the model.
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        layer_norm_epsilon (`float`, *optional*, defaults to 1e-5):
+            The epsilon used by the layer normalization layers.
+        residual_in_fp32 (`bool`, *optional*, defaults to `False`):
+            Whether or not residuals should be in `float32`. If set to `False` residuals will keep the same `dtype` as the rest of the model.
+        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`.
+        num_logits_to_keep (`int` or `None`, *optional*, defaults to 1):
+            Number of prompt logits to calculate during generation. If `None`, all logits will be calculated. If an
+            integer value, only last `num_logits_to_keep` logits will be calculated.
+        pad_token_id (`int`, *optional*, defaults to 0):
+            The id of the padding token.
+        bos_token_id (`int`, *optional*, defaults to 1):
+            The id of the "beginning-of-sequence" token.
+        eos_token_id (`int`, *optional*, defaults to 2):
+            The id of the "end-of-sequence" token.
+        sliding_window (`int`, *optional*, defaults to None):
+            Sliding window attention window size.
+        max_position_embeddings (`int`, *optional*, defaults to 4096):
+            The maximum sequence length that this model might ever be used with.
+        attention_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio for the attention probabilities.
+        hidden_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio for the hidden states.
+        use_mamba_kernels (`bool`, *optional*, defaults to `True`):
+            Flag indicating whether or not to use the fast mamba kernels. These are available only if `mamba-ssm` and
+            `causal-conv1d` are installed, and the mamba modules are running on a CUDA device.
+        ssm_state_size (`int`, *optional*, defaults to 128):
+            The dimension of the mamba state space latents.
+        mamba_num_heads (`int`, *optional*, defaults to 128):
+            Number of heads in Mamba layers.
+        mamba_n_groups (`int`, *optional*, defaults to 8):
+            Number of groups in Mamba layers.
+        mamba_head_dim (`int`, *optional*, defaults to 64):
+            Dimension of each Mamba head.
+        mamba_d_conv (`int`, *optional*, defaults to 4):
+            The size of the mamba convolution kernel.
+        mamba_expand (`int`, *optional*, defaults to 2):
+            Expanding factor used to determine the mamba intermediate size.
+        mamba_hidden_act (`str`, *optional*, defaults to "silu"):
+            The non-linear activation function in the Mamba layers.
+        mamba_dt_min (`float`, *optional*, defaults to 0.001):
+            Minimum value for the time step in Mamba.
+        mamba_dt_max (`float`, *optional*, defaults to 0.1):
+            Maximum value for the time step in Mamba.
+        mamba_dt_limit (`tuple`, *optional*, defaults to (0.0, float("inf"))):
+            Limits for the time step in Mamba.
+        mamba_dt_init_floor (`float`, *optional*, defaults to 1e-4):
+            Floor value for time step initialization in Mamba.
+        mamba_conv_bias (`bool`, *optional*, defaults to `True`):
+            Whether to use bias in the convolution layer of the mamba mixer block.
+        mamba_proj_bias (`bool`, *optional*, defaults to `False`):
+            Whether to use bias in the input and output projections of the mamba mixer block.
+        mamba_chunk_size (`int`, *optional*, defaults to 256):
+            Size of chunks for Mamba processing.
+        rescale_prenorm_residual (`bool`, *optional*, defaults to `True`):
+            Whether to rescale the pre-normalization residual connections.
+    """
+
+    model_type = "nemotron_h"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    def __init__(
+        self,
+        vocab_size=131072,
+        tie_word_embeddings=False,
+        hidden_size=4096,
+        intermediate_size=21504,
+        num_hidden_layers=52,
+        hybrid_override_pattern="M-M-M-M*-M-M-M-M-M*-M-M-M-M-M*-M-M-M-M-M*-M-M-M-M-M-",
+        num_attention_heads=32,
+        head_dim=128,
+        num_key_value_heads=8,  # nemo: num_query_groups
+        mlp_hidden_act="relu2",
+        attention_bias=False,
+        mlp_bias=False,
+        use_bias=False,
+        initializer_range=0.02, # nemo: init_method_std
+        layer_norm_epsilon=1e-5, # nemo: layernorm_epsilon
+        residual_in_fp32=False,  #  Megatron Core default value
+        use_cache=True,
+        num_logits_to_keep=1,
+        pad_token_id=0,
+        bos_token_id=1,
+        eos_token_id=2,
+        sliding_window=None,
+        max_position_embeddings=4096,
+        attention_dropout=0.0,
+        hidden_dropout=0.0, # * ADDED
+        use_mamba_kernels=True,
+        ssm_state_size=128, # mamba_state_size
+        mamba_num_heads=128,
+        mamba_n_groups=8,  # nemo: mamba_ssm_ngroups = num_heads
+        mamba_head_dim=64,
+        mamba_d_conv=4,
+        mamba_expand=2,
+        mamba_hidden_act="silu",
+        mamba_dt_min=0.001,
+        mamba_dt_max=0.1,
+        mamba_dt_limit=(0.0, float("inf")),
+        mamba_dt_init_floor=1e-4,
+        mamba_conv_bias=True,
+        mamba_proj_bias=False,
+        mamba_chunk_size=128,
+        rescale_prenorm_residual=True,
+        n_routed_experts=8,
+        n_shared_experts=1,
+        moe_intermediate_size=7688,
+        moe_shared_expert_intermediate_size=7688,
+        num_experts_per_tok=2,
+        routed_scaling_factor=1.0,
+        n_group=1,
+        topk_group=1,
+        norm_topk_prob=True,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.tie_word_embeddings = tie_word_embeddings
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.hybrid_override_pattern = hybrid_override_pattern
+        self.num_attention_heads = num_attention_heads
+        self.head_dim = head_dim
+        self.sliding_window = sliding_window
+        self.max_position_embeddings = max_position_embeddings
+        self.attention_dropout = attention_dropout
+        self.hidden_dropout = hidden_dropout
+
+        # Validate hybrid_override_pattern
+        # M: Mamba2, *: Attention, -: MLP
+        assert len(self.hybrid_override_pattern) == self.num_hidden_layers, "hybrid_override_pattern must have the same length as num_hidden_layers"
+        assert re.match(r"^[*-M]+$", self.hybrid_override_pattern), "hybrid_override_pattern must only contain characters 'M', '*', or '-'"
+
+        # 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.mlp_hidden_act = mlp_hidden_act
+        self.attention_bias = attention_bias
+        self.mlp_bias = mlp_bias
+        self.use_bias = use_bias
+        self.initializer_range = initializer_range
+        self.layer_norm_epsilon = layer_norm_epsilon
+        self.residual_in_fp32 = residual_in_fp32
+
+        self.use_cache = use_cache
+        self.num_logits_to_keep = num_logits_to_keep
+
+        self.use_mamba_kernels = use_mamba_kernels
+        self.n_groups = mamba_n_groups
+        self.mamba_head_dim = mamba_head_dim
+        self.ssm_state_size = ssm_state_size
+        self.mamba_num_heads = mamba_num_heads
+        self.conv_kernel = mamba_d_conv
+        self.expand = mamba_expand
+        self.mamba_hidden_act = mamba_hidden_act
+        self.time_step_min = mamba_dt_min
+        self.time_step_max = mamba_dt_max
+        self.time_step_limit = mamba_dt_limit
+        self.time_step_floor = mamba_dt_init_floor
+        self.use_conv_bias = mamba_conv_bias
+        self.mamba_proj_bias = mamba_proj_bias
+        self.chunk_size = mamba_chunk_size
+        self.rescale_prenorm_residual = rescale_prenorm_residual
+        self.n_routed_experts = n_routed_experts
+        self.n_shared_experts = n_shared_experts
+        self.moe_intermediate_size = moe_intermediate_size
+        self.moe_shared_expert_intermediate_size = moe_shared_expert_intermediate_size
+        self.num_experts_per_tok = num_experts_per_tok
+        self.routed_scaling_factor = routed_scaling_factor
+        self.n_group = n_group
+        self.topk_group = topk_group
+        self.norm_topk_prob = norm_topk_prob
+
+        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,
+        )
+
+    @property
+    def layers_block_type(self):
+        return [
+            "mamba" if self.hybrid_override_pattern[i] == "M" else
+            "attention" if self.hybrid_override_pattern[i] == "*" else
+            "mlp" if self.hybrid_override_pattern[i] == "-" else "moe"
+            for i in range(self.num_hidden_layers)]

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "do_sample": true,
+  "bos_token_id": 1,
+  "eos_token_id": [2, 11],
+  "pad_token_id": 0,
+  "temperature": 1.0,
+  "top_p": 1.0,
+  "transformers_version": "4.55.4"
+}

hf_quant_config.json

@@ -0,0 +1,73 @@
+{
+    "producer": {
+        "name": "modelopt",
+        "version": "0.29.0"
+    },
+    "quantization": {
+        "quant_algo": "FP8",
+        "kv_cache_quant_algo": "FP8",
+        "group_size": 16, 
+        "exclude_modules": [
+            "lm_head",
+            "backbone.layers.4.mixer.in_proj",
+            "backbone.layers.4.mixer.out_proj",
+            "backbone.layers.5.mixer.q_proj",
+            "backbone.layers.5.mixer.k_proj",
+            "backbone.layers.5.mixer.v_proj",
+            "backbone.layers.5.mixer.o_proj",
+            "backbone.layers.11.mixer.in_proj",
+            "backbone.layers.11.mixer.out_proj",
+            "backbone.layers.12.mixer.q_proj",
+            "backbone.layers.12.mixer.k_proj",
+            "backbone.layers.12.mixer.v_proj",
+            "backbone.layers.12.mixer.o_proj",
+            "backbone.layers.18.mixer.in_proj",
+            "backbone.layers.18.mixer.out_proj",
+            "backbone.layers.19.mixer.q_proj",
+            "backbone.layers.19.mixer.k_proj",
+            "backbone.layers.19.mixer.v_proj",
+            "backbone.layers.19.mixer.o_proj",
+            "backbone.layers.25.mixer.in_proj",
+            "backbone.layers.25.mixer.out_proj",
+            "backbone.layers.26.mixer.q_proj",
+            "backbone.layers.26.mixer.k_proj",
+            "backbone.layers.26.mixer.v_proj",
+            "backbone.layers.26.mixer.o_proj",
+            "backbone.layers.32.mixer.in_proj",
+            "backbone.layers.32.mixer.out_proj",
+            "backbone.layers.33.mixer.q_proj",
+            "backbone.layers.33.mixer.k_proj",
+            "backbone.layers.33.mixer.v_proj",
+            "backbone.layers.33.mixer.o_proj",
+            "backbone.layers.41.mixer.in_proj",
+            "backbone.layers.41.mixer.out_proj",
+            "backbone.layers.42.mixer.q_proj",
+            "backbone.layers.42.mixer.k_proj",
+            "backbone.layers.42.mixer.v_proj",
+            "backbone.layers.42.mixer.o_proj",
+            "backbone.layers.0.mixer.conv1d",
+            "backbone.layers.2.mixer.conv1d",
+            "backbone.layers.4.mixer.conv1d",
+            "backbone.layers.7.mixer.conv1d",
+            "backbone.layers.9.mixer.conv1d",
+            "backbone.layers.11.mixer.conv1d",
+            "backbone.layers.14.mixer.conv1d",
+            "backbone.layers.16.mixer.conv1d",
+            "backbone.layers.18.mixer.conv1d",
+            "backbone.layers.21.mixer.conv1d",
+            "backbone.layers.23.mixer.conv1d",
+            "backbone.layers.25.mixer.conv1d",
+            "backbone.layers.28.mixer.conv1d",
+            "backbone.layers.30.mixer.conv1d",
+            "backbone.layers.32.mixer.conv1d",
+            "backbone.layers.35.mixer.conv1d",
+            "backbone.layers.37.mixer.conv1d",
+            "backbone.layers.39.mixer.conv1d",
+            "backbone.layers.41.mixer.conv1d",
+            "backbone.layers.44.mixer.conv1d",
+            "backbone.layers.46.mixer.conv1d",
+            "backbone.layers.48.mixer.conv1d",
+            "backbone.layers.50.mixer.conv1d"
+        ]
+    }
+}

model-00001-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1faa3d94580c9b83d70dd6b7d2668b6ba0461e648ba5159f83e80139bf1369a7
+size 3999240760

model-00002-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7beb249f4dde47eb6199eacecb93724bd28b77126267ff4ba8bf292736a7ecb
+size 3997872560

model-00003-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a45b93f5da8f9e3029241a68a5eb88610178d703f6182853232995c82eddbf7
+size 3997873320

model-00004-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68a9d0c07d047f78758ccf2aed198c27061c6664cb0df2e8462e7a8753cccfcd
+size 3997873136

model-00005-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3850e4b8cd55248f1fd29d3d424fe9d5b77f8ce107ed7bd4f08dd6c980408bd8
+size 3997872896

model-00006-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f093ddc92fa57a634549cefe90c55f6c622432d3280cbcd3a12708aa814ab84
+size 3997872656

model-00007-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c39259e5dd9ada25e1fa94e237968c7c0f874ddb0aa456a0cdc42cdf1f5c7b19
+size 3997872376

model-00008-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fad59ac6cba1db25be784ce25c6cd356aa80a15f8a422c4470b7fc36bfc26aa0
+size 3997184888

model-00009-of-00010.json

@@ -0,0 +1,427 @@
+{
+    "metadata": {
+        "total_size": 698451040
+    },
+    "weight_map": {
+        "backbone.layers.51.mixer.experts.58.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.58.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.58.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.58.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.58.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.58.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.59.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.59.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.59.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.59.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.59.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.59.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.60.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.60.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.60.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.60.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.60.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.60.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.61.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.61.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.61.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.61.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.61.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.61.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.62.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.62.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.62.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.62.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.62.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.62.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.63.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.63.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.63.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.63.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.63.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.63.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.64.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.64.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.64.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.64.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.64.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.64.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.65.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.65.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.65.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.65.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.65.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.65.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.66.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.66.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.66.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.66.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.66.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.66.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.67.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.67.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.67.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.67.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.67.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.67.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.68.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.68.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.68.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.68.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.68.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.68.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.69.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.69.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.69.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.69.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.69.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.69.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.70.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.70.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.70.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.70.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.70.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.70.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.71.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.71.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.71.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.71.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.71.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.71.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.72.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.72.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.72.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.72.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.72.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.72.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.73.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.73.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.73.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.73.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.73.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.73.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.74.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.74.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.74.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.74.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.74.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.74.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.75.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.75.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.75.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.75.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.75.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.75.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.76.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.76.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.76.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.76.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.76.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.76.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.77.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.77.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.77.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.77.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.77.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.77.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.78.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.78.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.78.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.78.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.78.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.78.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.79.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.79.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.79.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.79.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.79.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.79.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.80.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.80.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.80.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.80.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.80.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.80.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.81.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.81.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.81.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.81.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.81.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.81.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.82.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.82.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.82.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.82.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.82.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.82.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.83.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.83.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.83.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.83.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.83.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.83.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.84.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.84.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.84.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.84.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.84.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.84.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.85.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.85.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.85.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.85.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.85.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.85.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.86.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.86.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.86.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.86.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.86.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.86.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.87.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.87.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.87.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.87.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.87.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.87.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.88.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.88.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.88.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.88.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.88.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.88.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.89.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.89.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.89.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.89.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.89.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.89.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.90.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.90.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.90.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.90.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.90.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.90.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.91.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.91.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.91.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.91.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.91.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.91.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.92.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.92.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.92.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.92.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.92.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.92.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.93.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.93.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.93.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.93.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.93.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.93.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.94.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.94.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.94.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.94.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.94.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.94.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.95.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.95.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.95.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.95.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.95.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.95.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.96.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.96.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.96.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.96.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.96.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.96.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.97.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.97.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.97.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.97.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.97.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.97.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.98.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.98.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.98.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.98.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.98.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.98.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.99.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.99.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.99.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.99.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.99.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.99.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.100.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.100.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.100.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.100.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.100.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.100.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.101.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.101.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.101.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.101.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.101.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.101.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.102.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.102.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.102.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.102.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.102.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.102.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.103.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.103.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.103.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.103.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.103.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.103.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.104.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.104.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.104.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.104.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.104.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.104.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.105.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.105.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.105.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.105.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.105.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.105.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.106.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.106.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.106.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.106.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.106.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.106.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.107.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.107.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.107.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.107.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.107.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.107.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.108.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.108.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.108.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.108.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.108.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.108.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.109.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.109.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.109.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.109.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.109.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.109.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.110.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.110.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.110.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.110.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.110.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.110.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.111.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.111.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.111.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.111.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.111.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.111.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.112.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.112.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.112.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.112.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.112.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.112.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.113.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.113.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.113.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.113.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.113.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.113.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.114.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.114.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.114.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.114.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.114.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.114.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.115.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.115.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.115.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.115.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.115.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.115.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.116.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.116.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.116.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.116.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.116.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.116.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.117.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.117.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.117.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.117.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.117.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.117.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.118.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.118.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.118.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.118.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.118.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.118.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.119.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.119.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.119.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.119.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.119.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.119.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.120.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.120.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.120.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.120.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.120.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.120.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.121.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.121.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.121.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.121.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.121.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.121.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.122.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.122.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.122.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.122.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.122.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.122.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.123.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.123.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.123.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.123.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.123.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.123.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.124.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.124.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.124.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.124.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.124.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.124.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.125.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.125.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.125.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.125.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.125.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.125.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.126.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.126.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.126.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.126.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.126.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.126.down_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.127.up_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.127.up_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.127.up_proj.input_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.127.down_proj.weight": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.127.down_proj.weight_scale": "model-00009-of-00010.safetensors",
+        "backbone.layers.51.mixer.experts.127.down_proj.input_scale": "model-00009-of-00010.safetensors"
+    }
+}

model-00009-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:220a0bb702f36c060c6fa6478a8c720fdba7a549b92b85929139e99fb3eb888b
+size 698500912

model-00010-of-00010.json

@@ -0,0 +1,6 @@
+{
+    "metadata": {
+        "total_size": 0
+    },
+    "weight_map": {}
+}

model-00010-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60d95b10b6e140a9626a7058d5038528f2ff80148dc4569b881db56052046509
+size 40

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6021eb6847e682f89aa52d5eb6e8c7d902a23acfc8137e25211cf84828f1592
+size 17077485