File size: 6,810 Bytes
c3fe4b2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
import math
from dataclasses import dataclass
import torch
import torch.nn as nn
from torch.nn import functional as F

class LlamaRMSNorm(nn.Module):
    def __init__(self, hidden_size, eps=1e-6):
        super().__init__()
        self.weight = nn.Parameter(torch.ones(hidden_size))
        self.eps = eps

    def forward(self, x):
        rms = torch.sqrt(torch.mean(x * x, dim=-1, keepdim=True) + self.eps)
        x_norm = x / rms
        return self.weight * x_norm

class LlamaRotaryEmbedding(nn.Module):
    def __init__(self, dim, max_position_embeddings=2048, base=10000):
        super().__init__()
        inv_freq = 1.0 / (base ** (torch.arange(0, dim, 2).float() / dim))
        self.register_buffer("inv_freq", inv_freq)
        self.max_position_embeddings = max_position_embeddings
        self.dim = dim

    def forward(self, x, seq_len):
        t = torch.arange(seq_len, device=x.device).type_as(self.inv_freq)
        freqs = torch.einsum("i,j->ij", t, self.inv_freq)
        emb = torch.cat((freqs, freqs), dim=-1)
        return emb

def rotate_half(x):
    x1, x2 = x[..., :x.shape[-1]//2], x[..., x.shape[-1]//2:]
    return torch.cat((-x2, x1), dim=-1)

def apply_rotary_pos_emb(q, k, cos, sin, position_ids):
    cos = cos.unsqueeze(0).unsqueeze(0)
    sin = sin.unsqueeze(0).unsqueeze(0)
    cos = cos.expand(q.shape[0], q.shape[1], -1, -1)
    sin = sin.expand(k.shape[0], k.shape[1], -1, -1)
    q_embed = (q * cos) + (rotate_half(q) * sin)
    k_embed = (k * cos) + (rotate_half(k) * sin)
    return q_embed, k_embed

class LlamaSdpaAttention(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.hidden_size = config.n_embd
        self.num_heads = config.n_head
        self.head_dim = config.n_embd // config.n_head
        self.num_key_value_heads = config.n_head // 3
        self.num_key_value_groups = self.num_heads // self.num_key_value_heads

        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.hidden_size, self.hidden_size, bias=False)
        self.rotary_emb = LlamaRotaryEmbedding(self.head_dim)

    def forward(self, x, attention_mask=None):
        B, T, C = x.size()
        q = self.q_proj(x).view(B, T, self.num_heads, self.head_dim)
        k = self.k_proj(x).view(B, T, self.num_key_value_heads, self.head_dim)
        v = self.v_proj(x).view(B, T, self.num_key_value_heads, self.head_dim)

        k = k.repeat_interleave(self.num_key_value_groups, dim=2)
        v = v.repeat_interleave(self.num_key_value_groups, dim=2)

        q = q.transpose(1, 2)
        k = k.transpose(1, 2)
        v = v.transpose(1, 2)

        rotary_emb = self.rotary_emb(x, T)
        cos, sin = rotary_emb.cos(), rotary_emb.sin()
        q, k = apply_rotary_pos_emb(q, k, cos, sin, None)

        out = F.scaled_dot_product_attention(q, k, v, is_causal=True)
        out = out.transpose(1, 2).contiguous().view(B, T, C)
        return self.o_proj(out)

class LlamaMLP(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.gate_proj = nn.Linear(config.n_embd, config.intermediate_size, bias=False)
        self.up_proj = nn.Linear(config.n_embd, config.intermediate_size, bias=False)
        self.down_proj = nn.Linear(config.intermediate_size, config.n_embd, bias=False)
        self.act_fn = nn.SiLU()

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

class LlamaDecoderLayer(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.input_layernorm = LlamaRMSNorm(config.n_embd)
        self.self_attn = LlamaSdpaAttention(config)
        self.post_attention_layernorm = LlamaRMSNorm(config.n_embd)
        self.mlp = LlamaMLP(config)

    def forward(self, x):
        residual = x
        x = self.input_layernorm(x)
        x = self.self_attn(x)
        x = residual + x

        residual = x
        x = self.post_attention_layernorm(x)
        x = self.mlp(x)
        x = residual + x
        return x

@dataclass
class SmolLM2Config:
    block_size: int = 2048
    vocab_size: int = 49152
    n_layer: int = 30
    n_head: int = 9
    n_embd: int = 576
    intermediate_size: int = 1536
    num_key_value_heads: int = 3
    rms_norm_eps: float = 1e-5
    rope_theta: float = 10000.0
    initializer_range: float = 0.041666666666666664
    use_cache: bool = True

class SmolLM2(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.config = config

        self.embed_tokens = nn.Embedding(config.vocab_size, config.n_embd)
        self.layers = nn.ModuleList([LlamaDecoderLayer(config) for _ in range(config.n_layer)])
        self.norm = LlamaRMSNorm(config.n_embd, eps=config.rms_norm_eps)
        self.lm_head = nn.Linear(config.n_embd, config.vocab_size, bias=False)
        self.embed_tokens.weight = self.lm_head.weight
        self.apply(self._init_weights)

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

    def forward(self, idx, targets=None):
        B, T = idx.size()
        x = self.embed_tokens(idx)
        
        for layer in self.layers:
            x = layer(x)
            
        x = self.norm(x)
        logits = self.lm_head(x)

        loss = None
        if targets is not None:
            loss = F.cross_entropy(logits.view(-1, logits.size(-1)), targets.view(-1))
            
        return logits, loss

    def generate(self, idx, max_new_tokens, temperature=1.0, top_k=None):
        for _ in range(max_new_tokens):
            idx_cond = idx[:, -self.config.block_size:]
            logits, _ = self(idx_cond)
            logits = logits[:, -1, :] / temperature
            
            if top_k is not None:
                v, _ = torch.topk(logits, min(top_k, logits.size(-1)))
                logits[logits < v[:, [-1]]] = float('-inf')
                
            probs = F.softmax(logits, dim=-1)
            idx_next = torch.multinomial(probs, num_samples=1)
            idx = torch.cat((idx, idx_next), dim=1)
        return idx