# coding=utf-8 # Copyright 2024 The Kyutai and HuggingFace Inc. teams. 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. from ...configuration_utils import PretrainedConfig class HeliumConfig(PretrainedConfig): r""" This is the configuration class to store the configuration of a [`HeliumModel`]. It is used to instantiate an Helium 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 Helium 2b model. e.g. [kyutai/helium-2b](https://huggingface.co/kyutai/helium-2b) 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 48000): Vocabulary size of the Helium model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`HeliumModel`] hidden_size (`int`, *optional*, defaults to 2560): Dimension of the hidden representations. intermediate_size (`int`, *optional*, defaults to 7040): Dimension of the MLP representations. num_hidden_layers (`int`, *optional*, defaults to 24): Number of hidden layers in the Transformer decoder. num_attention_heads (`int`, *optional*, defaults to 20): Number of attention heads for each attention layer in the Transformer decoder. num_key_value_heads (`int`, *optional*, defaults to 20): 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, check out [this paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to `num_attention_heads`. head_dim (`int`, *optional*, defaults to 128): The attention head dimension. hidden_act (`str` or `function`, *optional*, defaults to `"silu"`): The legacy activation function. It is overwritten by the `hidden_activation`. attention_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for the attention probabilities. max_position_embeddings (`int`, *optional*, defaults to 4096): 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-08): 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 to tie weight embeddings rope_theta (`float`, *optional*, defaults to 100000.0): The base period of the RoPE embeddings. pad_token_id (`int`, *optional*, defaults to 3): Padding token id. eos_token_id (`int` | `list`, *optional*, defaults to 2): End of stream token id. bos_token_id (`int`, *optional*, defaults to 1): Beginning of stream token id. attention_bias (`bool`, *optional*, defaults to `False`): Whether to use a bias in the query, key, value and output projection layers during self-attention. mlp_bias (`bool`, *optional*, defaults to `False`): Whether to use a bias in up_proj, down_proj and gate_proj layers in the MLP layers. ```python >>> from transformers import HeliumModel, HeliumConfig >>> # Initializing a Helium 2b style configuration >>> configuration = HeliumConfig() >>> # Initializing a model from the Helium 2b style configuration >>> model = HeliumModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```""" model_type = "helium" keys_to_ignore_at_inference = ["past_key_values"] base_model_tp_plan = { "layers.*.self_attn.q_proj": "colwise", "layers.*.self_attn.k_proj": "colwise", "layers.*.self_attn.v_proj": "colwise", "layers.*.self_attn.o_proj": "rowwise", "layers.*.mlp.gate_proj": "colwise", "layers.*.mlp.up_proj": "colwise", "layers.*.mlp.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=48000, hidden_size=2560, intermediate_size=7040, num_hidden_layers=24, num_attention_heads=20, num_key_value_heads=20, head_dim=128, hidden_act="silu", attention_dropout=0.0, max_position_embeddings=4096, initializer_range=0.02, rms_norm_eps=1e-8, use_cache=True, tie_word_embeddings=False, rope_theta=100000.0, pad_token_id=3, eos_token_id=2, bos_token_id=1, attention_bias=False, mlp_bias=False, **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.num_key_value_heads = num_key_value_heads self.head_dim = head_dim 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.attention_bias = attention_bias self.attention_dropout = attention_dropout self.mlp_bias = mlp_bias 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, ) __all__ = ["HeliumConfig"]