model.py

import math
from typing import Optional, Tuple, List
import torch
import torch.nn as nn
import torch.nn.functional as F

class RMSNorm(nn.Module):
    def __init__(self, dim: int, eps: float = 1e-5):
        super().__init__()
        self.eps = eps
        self.weight = nn.Parameter(torch.ones(dim))

    def forward(self, x):
        variance = x.pow(2).mean(-1, keepdim=True)
        x = x * torch.rsqrt(variance + self.eps)
        return self.weight * x

def rotate_half(x):
    x1, x2 = x.chunk(2, dim=-1)
    return torch.cat((-x2, x1), dim=-1)

class RotaryEmbedding(nn.Module):
    def __init__(self, dim: int, max_position_embeddings: int = 2048, base: int = 10000):
        super().__init__()
        inv_freq = 1.0 / (base ** (torch.arange(0, dim, 2).float() / dim))
        self.register_buffer("inv_freq", inv_freq)
        self.max_seq_len_cached = max_position_embeddings
        t = torch.arange(self.max_seq_len_cached).type_as(self.inv_freq)
        freqs = torch.einsum("i,j->ij", t, self.inv_freq)
        emb = torch.cat((freqs, freqs), dim=-1)
        self.register_buffer("cos_cached", emb.cos()[None, None, :, :])
        self.register_buffer("sin_cached", emb.sin()[None, None, :, :])

    def forward(self, x, seq_len=None):
        if seq_len > self.max_seq_len_cached:
            self.max_seq_len_cached = seq_len
            t = torch.arange(self.max_seq_len_cached).type_as(self.inv_freq)
            freqs = torch.einsum("i,j->ij", t, self.inv_freq)
            emb = torch.cat((freqs, freqs), dim=-1)
            self.register_buffer("cos_cached", emb.cos()[None, None, :, :])
            self.register_buffer("sin_cached", emb.sin()[None, None, :, :])
        return (
            self.cos_cached[:, :, :seq_len, ...],
            self.sin_cached[:, :, :seq_len, ...]
        )

def apply_rotary_pos_emb(q, k, cos, sin, position_ids=None):
    q_embed = (q * cos) + (rotate_half(q) * sin)
    k_embed = (k * cos) + (rotate_half(k) * sin)
    return q_embed, k_embed

class MLP(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.gate_proj = nn.Linear(config.hidden_size, config.intermediate_size, bias=False)
        self.up_proj = nn.Linear(config.hidden_size, config.intermediate_size, bias=False)
        self.down_proj = nn.Linear(config.intermediate_size, config.hidden_size, bias=False)
        self.act_fn = F.silu

    def forward(self, x):
        return self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))

class Attention(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.config = config
        self.hidden_size = config.hidden_size
        self.num_heads = config.num_attention_heads
        self.head_dim = config.hidden_size // config.num_attention_heads
        self.num_key_value_heads = config.num_key_value_heads
        self.max_position_embeddings = config.max_position_embeddings
        
        self.q_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=False)
        self.k_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=False)
        self.v_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=False)
        self.o_proj = nn.Linear(self.num_heads * self.head_dim, self.hidden_size, bias=False)
        
        self.rotary_emb = RotaryEmbedding(
            self.head_dim,
            max_position_embeddings=self.max_position_embeddings,
            base=config.rope_theta,
        )

    def forward(
        self,
        hidden_states: torch.Tensor,
        attention_mask: Optional[torch.Tensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        past_key_value: Optional[Tuple[torch.Tensor]] = None,
        output_attentions: bool = False,
        use_cache: bool = False,
    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
        bsz, q_len, _ = hidden_states.size()

        query_states = self.q_proj(hidden_states)
        key_states = self.k_proj(hidden_states)
        value_states = self.v_proj(hidden_states)

        query_states = query_states.view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
        key_states = key_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
        value_states = value_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)

        cos, sin = self.rotary_emb(value_states, seq_len=q_len)
        query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin, position_ids)

        if self.num_key_value_heads != self.num_heads:
            key_states = torch.repeat_interleave(key_states, self.num_heads // self.num_key_value_heads, dim=1)
            value_states = torch.repeat_interleave(value_states, self.num_heads // self.num_key_value_heads, dim=1)

        attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) / math.sqrt(self.head_dim)

        if attention_mask is not None:
            attn_weights = attn_weights + attention_mask
            
        attn_weights = F.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query_states.dtype)
        attn_output = torch.matmul(attn_weights, value_states)
        
        attn_output = attn_output.transpose(1, 2).contiguous()
        attn_output = attn_output.reshape(bsz, q_len, self.hidden_size)
        attn_output = self.o_proj(attn_output)

        return attn_output

class SmolLM2Block(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.hidden_size = config.hidden_size
        self.self_attn = Attention(config)
        self.mlp = MLP(config)
        self.input_layernorm = RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
        self.post_attention_layernorm = RMSNorm(config.hidden_size, eps=config.rms_norm_eps)

    def forward(
        self,
        hidden_states: torch.Tensor,
        attention_mask: Optional[torch.Tensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        past_key_value: Optional[Tuple[torch.Tensor]] = None,
        output_attentions: Optional[bool] = False,
        use_cache: Optional[bool] = False,
    ) -> Tuple[torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]]:
        
        residual = hidden_states
        hidden_states = self.input_layernorm(hidden_states)
        
        # Self Attention
        hidden_states = self.self_attn(
            hidden_states=hidden_states,
            attention_mask=attention_mask,
            position_ids=position_ids,
            past_key_value=past_key_value,
            output_attentions=output_attentions,
            use_cache=use_cache,
        )
        hidden_states = residual + hidden_states

        # MLP
        residual = hidden_states
        hidden_states = self.post_attention_layernorm(hidden_states)
        hidden_states = self.mlp(hidden_states)
        hidden_states = residual + hidden_states

        return hidden_states

class SmolLM2Model(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.config = config
        self.vocab_size = config.vocab_size
        self.hidden_size = config.hidden_size
        
        self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size)
        self.layers = nn.ModuleList([SmolLM2Block(config) for _ in range(config.num_hidden_layers)])
        self.norm = RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
        
        # Add gradient checkpointing flag
        self.gradient_checkpointing = False

        # Initialize weights
        self.apply(self._init_weights)

    def _init_weights(self, module):
        std = self.config.initializer_range
        if isinstance(module, nn.Linear):
            module.weight.data.normal_(mean=0.0, std=std)
            if module.bias is not None:
                module.bias.data.zero_()
        elif isinstance(module, nn.Embedding):
            module.weight.data.normal_(mean=0.0, std=std)

    def forward(
        self,
        input_ids: torch.LongTensor,
        attention_mask: Optional[torch.Tensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        past_key_values: Optional[List[torch.FloatTensor]] = None,
        inputs_embeds: Optional[torch.FloatTensor] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> torch.Tensor:
        
        if input_ids is not None:
            batch_size, seq_length = input_ids.shape
        else:
            batch_size, seq_length = inputs_embeds.shape[:2]

        if position_ids is None:
            device = input_ids.device if input_ids is not None else inputs_embeds.device
            position_ids = torch.arange(seq_length, dtype=torch.long, device=device)
            position_ids = position_ids.unsqueeze(0).expand(batch_size, -1)

        if inputs_embeds is None:
            inputs_embeds = self.embed_tokens(input_ids)

        hidden_states = inputs_embeds

        def create_custom_forward(module):
            def custom_forward(*inputs):
                return module(*inputs)
            return custom_forward

        if self.gradient_checkpointing and self.training:
            for layer in self.layers:
                hidden_states = torch.utils.checkpoint.checkpoint(
                    create_custom_forward(layer),
                    hidden_states,
                    attention_mask,
                    position_ids,
                    None,  # past_key_value
                    False,  # output_attentions
                    False,  # use_cache
                )
        else:
            for layer in self.layers:
                hidden_states = layer(
                    hidden_states,
                    attention_mask=attention_mask,
                    position_ids=position_ids,
                    past_key_value=None,
                    output_attentions=output_attentions,
                    use_cache=use_cache,
                )

        hidden_states = self.norm(hidden_states)

        return hidden_states

class SmolLM2ForCausalLM(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.config = config
        self.model = SmolLM2Model(config)
        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
        
        # Initialize weights and apply final processing
        self.post_init()

    def post_init(self):
        if self.config.tie_word_embeddings:
            self.lm_head.weight = self.model.embed_tokens.weight

    def forward(
        self,
        input_ids: torch.LongTensor,
        attention_mask: Optional[torch.Tensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        past_key_values: Optional[List[torch.FloatTensor]] = None,
        inputs_embeds: Optional[torch.FloatTensor] = None,
        labels: Optional[torch.LongTensor] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> torch.Tensor:
        hidden_states = self.model(
            input_ids=input_ids,
            attention_mask=attention_mask,
            position_ids=position_ids,
            past_key_values=past_key_values,
            inputs_embeds=inputs_embeds,
            use_cache=use_cache,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

        logits = self.lm_head(hidden_states)

        loss = None
        if labels is not None:
            shift_logits = logits[..., :-1, :].contiguous()
            shift_labels = labels[..., 1:].contiguous()
            loss = F.cross_entropy(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1))

        return logits, loss