File size: 7,603 Bytes
bd21ba5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
176
177
178
179
180
181
182
183
184
185
186
"""
MiniMind Max2 Main Model
Complete implementation of the Max2 language model.
"""

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

import sys
from pathlib import Path
sys.path.insert(0, str(Path(__file__).parent.parent))
from configs.model_config import Max2Config, get_config
from .components import Max2DecoderLayer, Max2RMSNorm


class Max2Model(nn.Module):
    """Max2 Transformer Model - outputs raw hidden states."""

    def __init__(self, config: Max2Config):
        super().__init__()
        self.config = config
        self.padding_idx = config.pad_token_id
        self.vocab_size = config.vocab_size

        self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, padding_idx=self.padding_idx)
        self.layers = nn.ModuleList([Max2DecoderLayer(config, i) for i in range(config.num_hidden_layers)])
        self.norm = Max2RMSNorm(config.hidden_size, eps=config.rms_norm_eps)

        self.gradient_checkpointing = False
        self._init_weights()

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

    def _make_causal_mask(self, seq_len: int, dtype: torch.dtype, device: torch.device) -> torch.Tensor:
        mask = torch.full((seq_len, seq_len), float("-inf"), dtype=dtype, device=device)
        mask = torch.triu(mask, diagonal=1)
        return mask.unsqueeze(0).unsqueeze(0)

    def forward(
        self,
        input_ids: torch.LongTensor,
        attention_mask: Optional[torch.Tensor] = None,
        past_key_values: Optional[List[Tuple[torch.Tensor, torch.Tensor]]] = None,
        use_cache: bool = False,
    ) -> Tuple[torch.Tensor, Optional[List[Tuple[torch.Tensor, torch.Tensor]]], torch.Tensor]:
        batch_size, seq_len = input_ids.shape
        hidden_states = self.embed_tokens(input_ids)

        causal_mask = self._make_causal_mask(seq_len, hidden_states.dtype, hidden_states.device)
        if attention_mask is not None:
            padding_mask = (1.0 - attention_mask[:, None, None, :].to(hidden_states.dtype)) * float("-inf")
            causal_mask = causal_mask + padding_mask

        next_cache = [] if use_cache else None
        total_aux_loss = torch.tensor(0.0, device=hidden_states.device)

        for idx, layer in enumerate(self.layers):
            past_kv = past_key_values[idx] if past_key_values else None
            hidden_states, present_kv, aux_loss = layer(hidden_states, causal_mask, past_kv, use_cache)

            if use_cache:
                next_cache.append(present_kv)
            total_aux_loss = total_aux_loss + aux_loss

        hidden_states = self.norm(hidden_states)
        return hidden_states, next_cache, total_aux_loss


class Max2ForCausalLM(nn.Module):
    """Max2 Model with Language Modeling head for text generation."""

    def __init__(self, config: Max2Config):
        super().__init__()
        self.config = config
        self.model = Max2Model(config)
        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
        self.lm_head.weight = self.model.embed_tokens.weight

    def forward(
        self,
        input_ids: torch.LongTensor,
        attention_mask: Optional[torch.Tensor] = None,
        labels: Optional[torch.LongTensor] = None,
        past_key_values: Optional[List[Tuple[torch.Tensor, torch.Tensor]]] = None,
        use_cache: bool = False,
    ) -> Tuple[Optional[torch.Tensor], torch.Tensor, Optional[List], torch.Tensor]:
        hidden_states, next_cache, aux_loss = self.model(input_ids, attention_mask, past_key_values, use_cache)
        logits = self.lm_head(hidden_states).float()

        loss = None
        if labels is not None:
            shift_logits = logits[..., :-1, :].contiguous()
            shift_labels = labels[..., 1:].contiguous()
            loss = CrossEntropyLoss()(shift_logits.view(-1, self.config.vocab_size), shift_labels.view(-1))
            loss = loss + aux_loss

        return loss, logits, next_cache, aux_loss

    @torch.no_grad()
    def generate(
        self,
        input_ids: torch.LongTensor,
        max_new_tokens: int = 100,
        temperature: float = 1.0,
        top_k: int = 50,
        top_p: float = 0.95,
        do_sample: bool = True,
    ) -> torch.LongTensor:
        """Simple generation with top-k/top-p sampling."""
        generated = input_ids
        past_key_values = None

        for _ in range(max_new_tokens):
            if past_key_values is None:
                _, logits, past_key_values, _ = self(generated, use_cache=True)
            else:
                _, logits, past_key_values, _ = self(generated[:, -1:], past_key_values=past_key_values, use_cache=True)

            next_token_logits = logits[:, -1, :] / temperature

            if do_sample:
                if top_k > 0:
                    indices_to_remove = next_token_logits < torch.topk(next_token_logits, top_k)[0][..., -1, None]
                    next_token_logits[indices_to_remove] = float('-inf')

                if top_p < 1.0:
                    sorted_logits, sorted_indices = torch.sort(next_token_logits, descending=True)
                    cumulative_probs = torch.cumsum(F.softmax(sorted_logits, dim=-1), dim=-1)
                    sorted_indices_to_remove = cumulative_probs > top_p
                    sorted_indices_to_remove[..., 1:] = sorted_indices_to_remove[..., :-1].clone()
                    sorted_indices_to_remove[..., 0] = 0
                    indices_to_remove = sorted_indices_to_remove.scatter(1, sorted_indices, sorted_indices_to_remove)
                    next_token_logits[indices_to_remove] = float('-inf')

                probs = F.softmax(next_token_logits, dim=-1)
                next_token = torch.multinomial(probs, num_samples=1)
            else:
                next_token = torch.argmax(next_token_logits, dim=-1, keepdim=True)

            generated = torch.cat([generated, next_token], dim=1)

            if (next_token == self.config.eos_token_id).all():
                break

        return generated


# Backward compatibility aliases
Mind2Model = Max2Model
Mind2ForCausalLM = Max2ForCausalLM


def create_model(model_name: str = "max2-lite", device: str = "cuda", dtype: torch.dtype = torch.float16) -> Max2ForCausalLM:
    """Factory function to create a Max2 model."""
    config = get_config(model_name)
    model = Max2ForCausalLM(config)
    return model.to(device=device, dtype=dtype) if torch.cuda.is_available() else model


if __name__ == "__main__":
    for model_name in ["max2-nano", "max2-lite", "max2-pro"]:
        print(f"\n{'='*50}\nTesting {model_name}\n{'='*50}")
        config = get_config(model_name)
        model = Max2ForCausalLM(config)

        total_params = sum(p.numel() for p in model.parameters())
        print(f"Total Parameters: {total_params / 1e9:.3f}B")

        input_ids = torch.randint(0, config.vocab_size, (2, 128))
        model.eval()
        with torch.no_grad():
            loss, logits, _, aux_loss = model(input_ids, labels=input_ids)
        print(f"Logits shape: {logits.shape}")
        print(f"Loss: {loss:.4f}, Aux loss: {aux_loss:.6f}")
        print("Forward pass successful!")