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 +1,157 @@
+{%- macro visible_text(content) -%}
+    {%- if content is string -%}
+        {{- content -}}
+    {%- elif content is iterable and content is not mapping -%}
+        {%- for item in content -%}
+            {%- if item is mapping and item.type == 'text' -%}
+                {{- item.text -}}
+            {%- elif item is string -%}
+                {{- item -}}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- else -%}
+        {{- content -}}
+    {%- endif -%}
+{%- endmacro -%}
+
+{{- '<|beginoftext|>' -}}
+
+{%- set ns = namespace(has_system=false, last_user_index=-1, last_assistant_index=-1) -%}
+{%- if messages | length > 0 and messages[0].role == 'system' -%}
+    {%- set ns.has_system = true -%}
+{%- endif -%}
+{%- for m in messages -%}
+    {%- if m.role == 'user' -%}
+        {%- set ns.last_user_index = loop.index0 -%}
+    {%- elif m.role == 'assistant' -%}
+        {%- set ns.last_assistant_index = loop.index0 -%}
+    {%- endif -%}
+{%- endfor -%}
+
+{#- ── System / Tools block ── -#}
+{%- if tools is iterable and tools | length > 0 -%}
+    {{- '<|startofturn|><|system|>' -}}
+    {{- '# Tools\n\nYou may call one or more functions to assist with the user query.\n\n' -}}
+    {{- 'You are provided with function signatures within <tools></tools> XML tags:\n\n<tools>' -}}
+    {%- for tool in tools -%}
+        {%- if tool.function is defined -%}
+            {%- set tool = tool.function -%}
+        {%- endif -%}
+        {{- '\n' ~ (tool | tojson) -}}
+    {%- endfor -%}
+    {{- '\n</tools>' -}}
+    {{- '\n\nFor each function call, output in JSON within <tool_call> tags:\n' -}}
+    {%- for tool in tools -%}
+        {%- if tool.function is defined -%}
+            {%- set tool = tool.function -%}
+        {%- endif -%}
+        {%- set _props = tool.parameters.properties if (tool.parameters is defined and tool.parameters.properties is defined) else {} -%}
+        {{- '\n<tool_call>{"name": "' ~ tool.name ~ '", "arguments": {' -}}
+        {%- set _keys = _props | list -%}
+        {%- for k in _keys -%}
+            {{- '"' ~ k ~ '": <' ~ k ~ '>' -}}
+            {%- if not loop.last -%}{{- ', ' -}}{%- endif -%}
+        {%- endfor -%}
+        {{- '}}</tool_call>' -}}
+    {%- endfor -%}
+    {%- if ns.has_system -%}
+        {{- '\n\n' ~ visible_text(messages[0].content) -}}
+    {%- endif -%}
+    {{- '<|endofturn|>' -}}
+{%- elif ns.has_system -%}
+    {{- '<|startofturn|><|system|>' ~ visible_text(messages[0].content) ~ '<|endofturn|>' -}}
+{%- endif -%}
+
+{#- ── Conversation turns ── -#}
+{%- for m in messages -%}
+
+    {#- [Fix 1] continue 대신 if/elif 체인으로 첫 system 스킵 -#}
+    {%- if loop.index0 == 0 and m.role == 'system' -%}
+        {#- already rendered above, skip -#}
+
+    {#- [Fix 2] 중간에 나오는 system 메시지도 처리 -#}
+    {%- elif m.role == 'system' -%}
+        {{- '<|startofturn|><|system|>' ~ visible_text(m.content) ~ '<|endofturn|>' -}}
+
+    {%- elif m.role == 'user' -%}
+        {{- '<|startofturn|><|user|>' -}}
+        {%- if m.references is defined and m.references -%}
+            {{- '<|reference|>' ~ m.references ~ '\n' -}}
+        {%- endif -%}
+        {{- visible_text(m.content) -}}
+        {{- '<|endofturn|>' -}}
+
+    {%- elif m.role == 'assistant' -%}
+        {{- '<|startofturn|><|assistant|>' -}}
+
+        {#- [순서 정책] think → plan → content → tool_calls -#}
+
+        {#- Reasoning block -#}
+        {%- set _content = visible_text(m.content) -%}
+        {%- set _reasoning = '' -%}
+        {%- if m.reasoning_content is string -%}
+            {%- set _reasoning = m.reasoning_content -%}
+        {%- elif '</think>' in _content -%}
+            {%- set _reasoning = _content.split('</think>')[0].split('<think>')[-1].strip() -%}
+            {%- set _content = _content.split('</think>', 1)[-1].lstrip('\n') -%}
+        {%- endif -%}
+        {%- set _has_tools = (tools is defined and tools is iterable and tools | length > 0) -%}
+        {%- set _emit_think = _reasoning and (_has_tools or loop.index0 == ns.last_assistant_index) -%}
+        {%- if _emit_think -%}
+            {{- '<think>' ~ _reasoning.strip() ~ '</think>' -}}
+        {%- endif -%}
+
+        {#- Text content -#}
+        {%- if _content.strip() -%}
+            {{- _content.strip() -}}
+        {%- endif -%}
+
+        {#- Tool calls — IDs are rendered for intermediate assistant turns
+             (context for call↔response correlation) but omitted for the last
+             assistant turn (prediction target — model should not learn to
+             generate IDs). After multi-turn data expansion, intermediate turns
+             become GRAY context and the last turn is GREEN. -#}
+        {%- if m.tool_calls is defined and m.tool_calls -%}
+            {%- set _is_last_assistant = (loop.index0 == ns.last_assistant_index) and not add_generation_prompt -%}
+            {%- for tc in m.tool_calls -%}
+                {%- set _tc_id = tc.id if (tc.id is defined and not _is_last_assistant) else none -%}
+                {%- if tc.function is defined -%}
+                    {%- set tc = tc.function -%}
+                {%- endif -%}
+                {%- set _id_suffix = ', "id": ' ~ (_tc_id | tojson) if _tc_id is not none else '' -%}
+                {%- if tc.arguments is not defined or not tc.arguments or tc.arguments == "" -%}
+                    {{- '\n<tool_call>' ~ '{"name": "' ~ tc.name ~ '", "arguments": ' ~ null ~ _id_suffix ~ '}' ~ '</tool_call>' -}}
+                {%- elif tc.arguments is string -%}
+                    {{- '\n<tool_call>' ~ '{"name": "' ~ tc.name ~ '", "arguments": ' ~ tc.arguments ~ _id_suffix ~ '}' ~ '</tool_call>' -}}
+                {%- else -%}
+                    {{- '\n<tool_call>' ~ '{"name": "' ~ tc.name ~ '", "arguments": ' ~ (tc.arguments | tojson) ~ _id_suffix ~ '}' ~ '</tool_call>' -}}
+                {%- endif -%}
+            {%- endfor -%}
+        {%- endif -%}
+
+        {{- '<|endofturn|>' -}}
+
+    {%- elif m.role == 'tool' -%}
+        {#- [Fix 3] loop.previtem/nextitem으로 인덱스 오버플로 제거 -#}
+        {%- if loop.first or loop.previtem.role != 'tool' -%}
+            {{- '<|startofturn|><|tool|>' -}}
+        {%- endif -%}
+        {{- '<tool_response>' ~ ({"tool_call_id": m.tool_call_id, "content": m.content} | tojson) ~ '</tool_response>' -}}
+        {%- if loop.last or loop.nextitem.role != 'tool' -%}
+            {{- '<|endofturn|>' -}}
+        {%- endif -%}
+
+    {%- endif -%}
+{%- endfor -%}
+
+{#- ── Generation prompt ── -#}
+{%- if add_generation_prompt -%}
+    {{- '<|startofturn|><|assistant|>' -}}
+    {%- if enable_thinking is defined and not enable_thinking -%}
+        {{- '<think></think>' -}}
+    {%- else -%}
+        {{- '<think>' -}}
+    {%- endif -%}
+{%- else -%}
+    {{- '<|endoftext|>' -}}
+{%- endif -%}

config.json

@@ -0,0 +1,81 @@
+{
+  "_debug_force_load_balance": false,
+  "architectures": [
+    "MotifForCausalLM"
+  ],
+  "attention_cls": "gdla",
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_motif.MotifConfig",
+    "AutoModel": "modeling_motif.MotifForCausalLM",
+    "AutoModelForCausalLM": "modeling_motif.MotifForCausalLM"
+  },
+  "diff_v2": true,
+  "dtype": "bfloat16",
+  "elementwise_attn_output_gate": true,
+  "eos_token_id": 0,
+  "experts_top_k": 8,
+  "head_dim": 192,
+  "headwise_attn_output_gate": false,
+  "hidden_act": "poly_norm",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "interleave_moe_layer_step": 1,
+  "intermediate_size": 12288,
+  "k_ratio": 1,
+  "kv_lora_rank": 512,
+  "load_balance_coeff": 0.0001,
+  "max_position_embeddings": 262144,
+  "max_window_layers": 9,
+  "mhc_enabled": true,
+  "mhc_expansion_rate": 4,
+  "mhc_identity_init": false,
+  "mhc_sinkhorn_iters": 20,
+  "model_type": "Motif",
+  "moe_intermediate_size": 1280,
+  "mscale": 1.0,
+  "n_dense_first_layers": 2,
+  "num_attention_heads": 80,
+  "num_experts": 384,
+  "num_hidden_layers": 53,
+  "num_key_value_heads": 16,
+  "num_noise_heads": 16,
+  "num_shared_experts": 1,
+  "output_router_logits": false,
+  "q_lora_rank": 1024,
+  "qk_rope_head_dim": 64,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "route_norm": true,
+  "route_scale": 2.0,
+  "router_aux_loss_coef": 0.0,
+  "score_before_experts": false,
+  "score_func": "sigmoid",
+  "sliding_window": 128,
+  "sliding_window_pattern": "interleave",
+  "sliding_window_period": 4,
+  "swa_rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "transformers_version": "5.7.0",
+  "use_cache": true,
+  "use_sliding_window": true,
+  "v_head_dim": 128,
+  "vocab_size": 220160,
+  "rope_factor": 64.0,
+  "original_seq_len": 4096,
+  "rope_scaling": {
+    "original_max_position_embeddings": 4096,
+    "factor": 64.0,
+    "mscale": 1.0,
+    "rope_type": "yarn",
+    "rope_theta": 10000.0,
+    "beta_fast": 32.0,
+    "beta_slow": 1.0,
+    "apply_yarn_scaling": false
+  },
+  "polynorm_output_scale": 0.5,
+  "polynorm_output_scale_per_layer": {},
+  "polynorm_bias_clamp": 0.5,
+  "hidden_clamp": 1000000.0,
+  "num_nextn_predict_layers": 1
+}

configuration_motif.py

@@ -0,0 +1,294 @@
+import torch
+from transformers.configuration_utils import PretrainedConfig
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+
+class MotifConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`MotifModel`]. It is used to instantiate a
+    Motif model according to the specified arguments, defining the model architecture.
+    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 151936):
+            Vocabulary size of the Motif model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`MotifModel`]
+        hidden_size (`int`, *optional*, defaults to 4096):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 22016):
+            Dimension of the MLP representations.
+        num_hidden_layers (`int`, *optional*, defaults to 32):
+            Number of hidden layers in the Transformer encoder.
+        num_attention_heads (`int`, *optional*, defaults to 32):
+            Number of attention heads for each attention layer in the Transformer encoder.
+        num_key_value_heads (`int`, *optional*, defaults to 32):
+            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 `32`.
+        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 32768):
+            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`.
+        tie_word_embeddings (`bool`, *optional*, defaults to `False`):
+            Whether the model's input and output word embeddings should be tied.
+        rope_theta (`float`, *optional*, defaults to 1000000.0):
+            The base period of the RoPE embeddings.
+        rope_scaling (`Dict`, *optional*):
+            Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type
+            and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value
+            accordingly.
+            Expected contents:
+                `rope_type` (`str`):
+                    The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope',
+                    'llama3'], with 'default' being the original RoPE implementation.
+                `factor` (`float`, *optional*):
+                    Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings. In
+                    most scaling types, a `factor` of x will enable the model to handle sequences of length x *
+                    original maximum pre-trained length.
+                `original_max_position_embeddings` (`int`, *optional*):
+                    Used with 'dynamic', 'longrope' and 'llama3'. The original max position embeddings used during
+                    pretraining.
+                `attention_factor` (`float`, *optional*):
+                    Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention
+                    computation. If unspecified, it defaults to value recommended by the implementation, using the
+                    `factor` field to infer the suggested value.
+                `beta_fast` (`float`, *optional*):
+                    Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear
+                    ramp function. If unspecified, it defaults to 32.
+                `beta_slow` (`float`, *optional*):
+                    Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear
+                    ramp function. If unspecified, it defaults to 1.
+                `short_factor` (`List[float]`, *optional*):
+                    Only used with 'longrope'. The scaling factor to be applied to short contexts (<
+                    `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
+                    size divided by the number of attention heads divided by 2
+                `long_factor` (`List[float]`, *optional*):
+                    Only used with 'longrope'. The scaling factor to be applied to long contexts (<
+                    `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
+                    size divided by the number of attention heads divided by 2
+                `low_freq_factor` (`float`, *optional*):
+                    Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE
+                `high_freq_factor` (`float`, *optional*):
+                    Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE
+        use_sliding_window (`bool`, *optional*, defaults to `False`):
+            Whether to use sliding window attention.
+        sliding_window (`int`, *optional*, defaults to 4096):
+            Sliding window attention (SWA) window size. If not specified, will default to `4096`.
+        max_window_layers (`int`, *optional*, defaults to 28):
+            The number of layers that use SWA (Sliding Window Attention). The bottom layers use SWA while the top use full attention.
+        attention_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio for the attention probabilities.
+    ```python
+    >>> from transformers import MotifModel, MotifConfig
+    >>> # Initializing a Motif style configuration
+    >>> configuration = MotifConfig()
+    >>> # Initializing a model from the Motif-102B style configuration
+    >>> model = MotifModel(configuration)
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```"""
+
+    model_type = "Motif"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    base_model_tp_plan = {
+        # Attention
+        "layers.*.self_attn.q_proj": "colwise",
+        "layers.*.self_attn.k_proj": "colwise",
+        "layers.*.self_attn.v_proj": "colwise",
+        "layers.*.self_attn.o_proj": "rowwise",
+        # Dense MLP
+        "layers.*.mlp.gate_proj": "colwise",
+        "layers.*.mlp.up_proj": "colwise",
+        "layers.*.mlp.down_proj": "rowwise",
+        # MoE experts (fused gate+up)
+        "layers.*.moe.experts.gate_up_proj": "packed_colwise",
+        "layers.*.moe.experts.down_proj": "rowwise",
+        # Shared experts
+        "layers.*.moe.shared_experts.gate_proj": "colwise",
+        "layers.*.moe.shared_experts.up_proj": "colwise",
+        "layers.*.moe.shared_experts.down_proj": "rowwise",
+    }
+
+    base_model_pp_plan = {
+        "embed_tokens": (["input_ids"], ["inputs_embeds"]),
+        "layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
+        "norm": (["hidden_states"], ["hidden_states"]),
+    }
+
+    def __init__(
+        self,
+        vocab_size=151936,
+        hidden_size=4096,
+        intermediate_size=22016,
+        num_hidden_layers=32,
+        num_attention_heads=32,
+        num_key_value_heads=32,
+        hidden_act="silu",
+        max_position_embeddings=32768,
+        initializer_range=0.02,
+        rms_norm_eps=1e-6,
+        use_cache=True,
+        tie_word_embeddings=False,
+        rope_theta=1000000.0,
+        rope_scaling=None,
+        use_sliding_window=False,
+        sliding_window=4096,
+        max_window_layers=28,
+        sliding_window_pattern="interleave",
+        sliding_window_period=2,
+        attention_dropout=0.0,
+        # Differential Attention parameters
+        head_dim=None,
+        num_noise_heads=0,
+        k_ratio=1,
+        # MoE parameters
+        num_experts=0,
+        experts_top_k=2,
+        num_shared_experts=0,
+        interleave_moe_layer_step=0,
+        moe_intermediate_size=None,
+        score_func="softmax",
+        route_norm=False,
+        route_scale=1.0,
+        load_balance_coeff=None,
+        score_before_experts=False,
+        _debug_force_load_balance=False,
+        output_router_logits=False,
+        router_aux_loss_coef=0.0,
+        # MHC (Manifold-constrained Hyper-Connections) parameters
+        mhc_enabled=False,
+        mhc_expansion_rate=4,
+        mhc_identity_init=False,
+        mhc_sinkhorn_iters=20,
+        # DiffAttention V2 / Attention class
+        diff_v2=False,
+        attention_cls="basic",
+        # GDLA (Grouped Differential Latent Attention) parameters
+        q_lora_rank=0,
+        kv_lora_rank=0,
+        qk_rope_head_dim=None,
+        v_head_dim=None,
+        original_seq_len=32768,
+        rope_factor=1.0,
+        mscale=1.0,
+        swa_rope_theta=None,
+        # Attention output gating
+        headwise_attn_output_gate=False,
+        elementwise_attn_output_gate=False,
+        # MoE: first N layers always dense (no MoE), regardless of interleave schedule
+        n_dense_first_layers=0,
+        # MTP (Multi-Token Prediction) speculative decoding
+        num_nextn_predict_layers=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.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.use_sliding_window = use_sliding_window
+        self.sliding_window = sliding_window if use_sliding_window else None
+        self.max_window_layers = max_window_layers
+        self.sliding_window_pattern = sliding_window_pattern
+        self.sliding_window_period = sliding_window_period
+
+        # 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.use_cache = use_cache
+        self.rope_theta = rope_theta
+        self.rope_scaling = rope_scaling
+        self.attention_dropout = attention_dropout
+
+        # Differential Attention configuration
+        self.head_dim = head_dim
+        self.num_noise_heads = num_noise_heads
+        self.k_ratio = k_ratio
+
+        # MoE configuration
+        self.num_experts = num_experts
+        self.experts_top_k = experts_top_k
+        self.num_shared_experts = num_shared_experts
+        self.interleave_moe_layer_step = interleave_moe_layer_step
+        self.moe_intermediate_size = moe_intermediate_size if moe_intermediate_size is not None else intermediate_size
+        self.score_func = score_func
+        self.route_norm = route_norm
+        self.route_scale = route_scale
+        self.load_balance_coeff = load_balance_coeff
+        self.score_before_experts = score_before_experts
+        self._debug_force_load_balance = _debug_force_load_balance
+        self.output_router_logits = output_router_logits
+        self.router_aux_loss_coef = router_aux_loss_coef
+
+        # MHC configuration
+        self.mhc_enabled = mhc_enabled
+        self.mhc_expansion_rate = mhc_expansion_rate
+        self.mhc_identity_init = mhc_identity_init
+        self.mhc_sinkhorn_iters = mhc_sinkhorn_iters
+
+        # DiffAttention V2 / Attention class
+        self.diff_v2 = diff_v2
+        self.attention_cls = attention_cls
+
+        # GDLA parameters
+        self.q_lora_rank = q_lora_rank
+        self.kv_lora_rank = kv_lora_rank
+        self.qk_rope_head_dim = qk_rope_head_dim
+        self.v_head_dim = v_head_dim
+        self.original_seq_len = original_seq_len
+        self.rope_factor = rope_factor
+        self.mscale = mscale
+        self.swa_rope_theta = swa_rope_theta
+
+        # Attention output gating
+        self.headwise_attn_output_gate = headwise_attn_output_gate
+        self.elementwise_attn_output_gate = elementwise_attn_output_gate
+
+        # MoE dense-first layers
+        self.n_dense_first_layers = n_dense_first_layers
+
+        # MTP speculative decoding
+        self.num_nextn_predict_layers = num_nextn_predict_layers
+
+        # Validate the correctness of rotary position embeddings parameters
+        # BC: if there is a 'type' field, move it to 'rope_type'.
+        if self.rope_scaling is not None and "type" in self.rope_scaling:
+            self.rope_scaling["rope_type"] = self.rope_scaling["type"]
+        if callable(getattr(type(self), "validate_rope", None)):
+            self.validate_rope()
+
+        # On ROCm, torch._grouped_mm is not supported at runtime even though
+        # transformers auto-selects the grouped_mm expert backend for torch>=2.9.
+        # Force eager (for-loop) dispatch so MoE models work on ROCm.
+        if (
+            self.num_experts > 0
+            and hasattr(torch.version, "hip")
+            and torch.version.hip is not None
+            and "experts_implementation" not in kwargs
+        ):
+            kwargs["experts_implementation"] = "eager"
+
+        super().__init__(
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )
+        logger.info(f" kwargs : {kwargs}")

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "pad_token_id": 0,
+  "eos_token_id": [0, 3, 6],
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "transformers_version": "5.7.0",
+  "use_cache": true,
+  "do_sample": true,
+  "temperature": 1.0,
+  "top_p": 0.95
+}

model-00001-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d448f19cf7d6e5deb7bc4eb33c11b6a29e3ba38ed06f36161739673d482325a
+size 2474208740

model-00002-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1eb1652398b1cd781e63ba7212ba871857896d927b49a9ff039aa5d450750c8
+size 336158060

model-00003-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:779acccca99ba6820889f6dfa9c975ccb0a8ef0f6850f6565543525d9f7e8b9b
+size 4026531992

model-00004-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efbdffbdc2a0b0f08b69c1b79dfdbacbc60d489d8b69d442d9b35103139ff20c
+size 169438244

model-00005-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a261c05852c76d8aea3f007595914b79616bafc5d8b610871248cbe6424c66d8
+size 4026531992

model-00006-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c735774006a755ba75deb7a111107133c746905878ce1682f2d6de9f66128cb
+size 169438244

model-00007-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2692b0a753bbf3b43eeef9c8bc10bb01324d0c1d49dd488d13362b0f3d836b2
+size 4026531992

model-00008-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da1c4e051658c8acd142485621d3e113d948d7caf99048c38b4f47d2a20fc837
+size 169438244

model-00009-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f660e9baa2f5caa5ffee4b3fdd5c734c87726c6f68a51f5bf36287fb05b7adce
+size 4026531992

model-00010-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08c389d5a08e7a7effef1c25d687f086dce81857b5ac51aef567514c9cb1b6f8
+size 169438244

model-00011-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb2d91468b652ad09c70665e4ed9a29557deddcbf7c6cd9a70f57a59b03d87f1
+size 4026531992

model-00012-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e36df01f27ef3e663552b27216fdb9a46955f4936b7e39cd246422a2a10731b
+size 169438244

model-00013-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ae9f93e85aa948b05c3e3030fddf1ed10c80300db4c65061da6b522fb6f056e
+size 4026531992

model-00014-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52b26ded92e96ee7dd743d35c977ffb2b2758bb5ee8a1d58f740009396ec2b45
+size 169438244

model-00015-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa0724f6c5426337f429855b4c59e0bf95d8da2fca7f44e509164bf1aa317d46
+size 4026531992

model-00016-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e58c2cf0ca136addd3f5c129ab6b5f2266c121a97a9256ed3198b9aad422f71
+size 169438244

model-00017-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:618b62820764ac2ee205816a4efefd28b8b3b7b5c5a1ffed4d2ca3959db64e4e
+size 4026531992

model-00018-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c20761c60352617d41c9277e9bd0cceabb69523ce538788c46fced0cf91ca75
+size 169438396

model-00019-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d57eaf08774e881ed4ebc286292f273384b0e2a24f58a320d47e49857c96b96a
+size 4026531992

model-00020-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83f8879330ec2bfb707e36d18b5183f5a588cc2a7e1cc813c0b8dc0cffb841b9
+size 169438284

model-00021-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:160ca8b2a98bc7b63a9107c7e6dafbb041bc109e668c16313374773a74401f5e
+size 4026531992

model-00022-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:950be62cc981471239196a31e1393c05b5c07710f8773d32f197b4002e3c111d
+size 169438284

model-00023-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ca72ad4dabe121f4fde880ae081355c921a616e13481ab4cb1d091e11d30ac8
+size 4026531992

model-00024-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d052825b6c7eb3d35d61bb1a7431989b84fa507a99b0c5eba9351d246178fac
+size 169438284

model-00025-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64471596c543df81d52d0bf691d4ca21b28ec19f03a78774725f58846bc636b3
+size 4026531992

model-00026-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:416e57726a13ba8715b6d5a5e13851ba95ae0166786883b7f50a71d9e204e6f4
+size 169438284

model-00027-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b27ccdf68027deb2d89b96733f44ae02c28205957951e1d5a0ccb19d15af05f
+size 4026531992

model-00028-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd66917ed5061a086da6053190b2484a49ec3012d3021ef75c03450b454886d5
+size 169438284

model-00029-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:353ad9a5ed504382fd80555ead5cb247dcf9ab52060b014274af8061b82b6b15
+size 4026531992

model-00030-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:449d4df572c901fd3c09afa890198bf10e96006788374475557dd958c8278cc9
+size 169438284

model-00031-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c599890131f8c3bec1d4c7683a724ed80ecc3f826246d5e9b60f11eb1ac490d9
+size 4026531992

model-00032-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e57ccd45eecf0809bacfd82d7ea216280206d1edc9ba48b4eba42b21be4ef2d
+size 169438284

model-00033-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4e22c1094e1ea7ca8a8ab4615e8ccc5b2210c58743c796147df0072dbeb22a1
+size 4026531992

model-00034-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa271bb63ac57361f39beb626681ea6ebb48fd78dd94a842d6588371adec8eb5
+size 169438284

model-00035-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bedbbdf120394fe75565200234475ea4f1b29928cdba87b3a9f8a593e06dc96d
+size 4026531992

model-00036-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2461806db82a88db31f2886ea5702c85e9e43e74b61941a3087c73fadbb92c34
+size 169438284

model-00037-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a077ad2d9b8924da94efb1dd6691e248ee4ad1271468409fd522611b792cead1
+size 4026531992

model-00038-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:989273cc306fd1e32c538fc2490db41de7927c31f71c444ab9c494ee68508ca3
+size 169438284

model-00039-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cb963f713ac084448b6fc767f15a9a3ca1b1318361f7030770afc4928f17cdb
+size 4026531992

model-00040-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8b8ab9ba70831205339498bfe49055f5194b9349f0e0cda1986b93b14fdcebb
+size 169438284

model-00041-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45fcfb10ee7bfd84baa379efdcfb62aab78fc190d1a4e837de9ea77be26d5d27
+size 4026531992

model-00042-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc71f5151978672b0f47de152faf30dd34628ac883c6800187ced6e09c81cfb1
+size 169438284

model-00043-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38f43a3fdb639d9af510f02a3d48c66668cbe6850c9e36897eeadbc0a28a52fa
+size 4026531992

model-00044-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b69854117c9f9d93fd4cb969b0b1a5e927bbe300a4530306e04335c1ccb04470
+size 169438284

model-00045-of-00155.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ea007d3929f682ab85db150dffc3f1cd9753dd2420fff791d37bcca9d9a26bf
+size 4026531992