530 lines
22 KiB
Python
530 lines
22 KiB
Python
import math
|
||
import struct
|
||
import inspect
|
||
from .LMConfig import LMConfig
|
||
from typing import Any, Optional, Tuple
|
||
import numpy as np
|
||
import torch
|
||
import torch.nn.functional as F
|
||
from torch import nn
|
||
from transformers import PreTrainedModel
|
||
from transformers.modeling_outputs import CausalLMOutputWithPast
|
||
|
||
|
||
class RMSNorm(torch.nn.Module):
|
||
def __init__(self, dim: int, eps: float):
|
||
super().__init__()
|
||
self.eps = eps
|
||
self.weight = nn.Parameter(torch.ones(dim))
|
||
|
||
def _norm(self, x):
|
||
return x * torch.rsqrt(x.pow(2).mean(-1, keepdim=True) + self.eps)
|
||
|
||
def forward(self, x):
|
||
output = self._norm(x.float()).type_as(x)
|
||
return output * self.weight
|
||
|
||
|
||
def precompute_pos_cis(dim: int, end: int, theta: float = 10000.0):
|
||
freqs = 1.0 / (theta ** (torch.arange(0, dim, 2)[: (dim // 2)].float() / dim))
|
||
t = torch.arange(end, device=freqs.device) # type: ignore
|
||
freqs = torch.outer(t, freqs).float() # type: ignore
|
||
pos_cis = torch.polar(torch.ones_like(freqs), freqs) # complex64
|
||
return pos_cis
|
||
|
||
|
||
def apply_rotary_emb(xq, xk, pos_cis):
|
||
def unite_shape(pos_cis, x):
|
||
ndim = x.ndim
|
||
assert 0 <= 1 < ndim
|
||
assert pos_cis.shape == (x.shape[1], x.shape[-1])
|
||
shape = [d if i == 1 or i == ndim - 1 else 1 for i, d in enumerate(x.shape)]
|
||
return pos_cis.view(*shape)
|
||
|
||
xq_ = torch.view_as_complex(xq.float().reshape(*xq.shape[:-1], -1, 2))
|
||
xk_ = torch.view_as_complex(xk.float().reshape(*xk.shape[:-1], -1, 2))
|
||
pos_cis = unite_shape(pos_cis, xq_)
|
||
xq_out = torch.view_as_real(xq_ * pos_cis).flatten(3)
|
||
xk_out = torch.view_as_real(xk_ * pos_cis).flatten(3)
|
||
return xq_out.type_as(xq), xk_out.type_as(xk)
|
||
|
||
|
||
def repeat_kv(x: torch.Tensor, n_rep: int) -> torch.Tensor:
|
||
"""torch.repeat_interleave(x, dim=2, repeats=n_rep)"""
|
||
bs, slen, n_kv_heads, head_dim = x.shape
|
||
if n_rep == 1:
|
||
return x
|
||
return (
|
||
x[:, :, :, None, :]
|
||
.expand(bs, slen, n_kv_heads, n_rep, head_dim)
|
||
.reshape(bs, slen, n_kv_heads * n_rep, head_dim)
|
||
)
|
||
|
||
|
||
class Attention(nn.Module):
|
||
def __init__(self, args: LMConfig):
|
||
super().__init__()
|
||
self.n_kv_heads = args.n_heads if args.n_kv_heads is None else args.n_kv_heads
|
||
assert args.n_heads % self.n_kv_heads == 0
|
||
model_parallel_size = 1
|
||
self.n_local_heads = args.n_heads // model_parallel_size
|
||
self.n_local_kv_heads = self.n_kv_heads // model_parallel_size
|
||
self.n_rep = self.n_local_heads // self.n_local_kv_heads
|
||
self.head_dim = args.dim // args.n_heads
|
||
self.wq = nn.Linear(args.dim, args.n_heads * self.head_dim, bias=False)
|
||
self.wk = nn.Linear(args.dim, self.n_kv_heads * self.head_dim, bias=False)
|
||
self.wv = nn.Linear(args.dim, self.n_kv_heads * self.head_dim, bias=False)
|
||
self.wo = nn.Linear(args.n_heads * self.head_dim, args.dim, bias=False)
|
||
self.attn_dropout = nn.Dropout(args.dropout)
|
||
self.resid_dropout = nn.Dropout(args.dropout)
|
||
self.dropout = args.dropout
|
||
|
||
# use flash attention or a manual implementation?
|
||
self.flash = hasattr(torch.nn.functional, 'scaled_dot_product_attention') and args.flash_attn
|
||
|
||
if not self.flash:
|
||
# print("WARNING: using slow attention. Flash Attention requires PyTorch >= 2.0")
|
||
mask = torch.full((1, 1, args.max_seq_len, args.max_seq_len), float("-inf"))
|
||
mask = torch.triu(mask, diagonal=1)
|
||
self.register_buffer("mask", mask)
|
||
|
||
def forward(
|
||
self,
|
||
x: torch.Tensor,
|
||
pos_cis: torch.Tensor,
|
||
use_kv_cache: bool = False,
|
||
past_kv: Tuple[torch.Tensor] = None
|
||
):
|
||
bsz, seqlen, _ = x.shape
|
||
# QKV
|
||
# inference
|
||
if use_kv_cache:
|
||
# 只计算最后一个token的Q
|
||
current_token = x[:, -1:, :]
|
||
|
||
if not past_kv:
|
||
xq = self.wq(x)
|
||
xk, xv = self.wk(x), self.wv(x)
|
||
else:
|
||
past_key, past_value = past_kv
|
||
xq = torch.cat((torch.zeros_like(x[:, :-1, :]), self.wq(current_token)), dim=1)
|
||
xk = torch.cat((past_key, self.wk(current_token)), dim=1)
|
||
xv = torch.cat((past_value, self.wv(current_token)), dim=1)
|
||
|
||
past_kv = (xk, xv)
|
||
else:
|
||
xq = self.wq(x)
|
||
xk, xv = self.wk(x), self.wv(x)
|
||
|
||
xq = xq.view(bsz, seqlen, self.n_local_heads, self.head_dim)
|
||
xk = xk.view(bsz, seqlen, self.n_local_kv_heads, self.head_dim)
|
||
xv = xv.view(bsz, seqlen, self.n_local_kv_heads, self.head_dim)
|
||
|
||
# RoPE relative positional embeddings
|
||
xq, xk = apply_rotary_emb(xq, xk, pos_cis)
|
||
|
||
# grouped multiquery attention: expand out keys and values
|
||
xk = repeat_kv(xk, self.n_rep) # (bs, seqlen, n_local_heads, head_dim)
|
||
xv = repeat_kv(xv, self.n_rep) # (bs, seqlen, n_local_heads, head_dim)
|
||
|
||
# make heads into a batch dimension
|
||
xq = xq.transpose(1, 2) # (bs, n_local_heads, seqlen, head_dim)
|
||
xk = xk.transpose(1, 2)
|
||
xv = xv.transpose(1, 2)
|
||
|
||
# flash implementation
|
||
if self.flash:
|
||
output = torch.nn.functional.scaled_dot_product_attention(xq, xk, xv, attn_mask=None,
|
||
dropout_p=self.dropout if self.training else 0.0,
|
||
is_causal=True)
|
||
else:
|
||
# manual implementation
|
||
scores = torch.matmul(xq, xk.transpose(2, 3)) / math.sqrt(self.head_dim)
|
||
assert hasattr(self, 'mask')
|
||
scores = scores + self.mask[:, :, :seqlen, :seqlen] # (bs, n_local_heads, seqlen, cache_len + seqlen)
|
||
scores = F.softmax(scores.float(), dim=-1).type_as(xq)
|
||
scores = self.attn_dropout(scores)
|
||
output = torch.matmul(scores, xv) # (bs, n_local_heads, seqlen, head_dim)
|
||
|
||
# restore time as batch dimension and concat heads
|
||
output = output.transpose(1, 2).contiguous().view(bsz, seqlen, -1)
|
||
|
||
# final projection into the residual stream
|
||
output = self.wo(output)
|
||
output = self.resid_dropout(output)
|
||
return output, past_kv
|
||
|
||
|
||
class FeedForward(nn.Module):
|
||
def __init__(self, dim: int, hidden_dim: int, multiple_of: int, dropout: float):
|
||
super().__init__()
|
||
if hidden_dim is None:
|
||
hidden_dim = 4 * dim
|
||
hidden_dim = int(2 * hidden_dim / 3)
|
||
hidden_dim = multiple_of * ((hidden_dim + multiple_of - 1) // multiple_of)
|
||
self.w1 = nn.Linear(dim, hidden_dim, bias=False)
|
||
self.w2 = nn.Linear(hidden_dim, dim, bias=False)
|
||
self.w3 = nn.Linear(dim, hidden_dim, bias=False)
|
||
self.dropout = nn.Dropout(dropout)
|
||
|
||
def forward(self, x):
|
||
return self.dropout(self.w2(F.silu(self.w1(x)) * self.w3(x)))
|
||
|
||
|
||
class MoEGate(nn.Module):
|
||
def __init__(self, config: LMConfig):
|
||
super().__init__()
|
||
self.config = config
|
||
self.top_k = config.num_experts_per_tok
|
||
self.n_routed_experts = config.n_routed_experts
|
||
|
||
self.scoring_func = config.scoring_func
|
||
self.alpha = config.aux_loss_alpha
|
||
self.seq_aux = config.seq_aux
|
||
|
||
# topk selection algorithm
|
||
self.norm_topk_prob = config.norm_topk_prob
|
||
self.gating_dim = config.dim
|
||
self.weight = nn.Parameter(torch.empty((self.n_routed_experts, self.gating_dim)))
|
||
self.reset_parameters()
|
||
|
||
def reset_parameters(self) -> None:
|
||
import torch.nn.init as init
|
||
init.kaiming_uniform_(self.weight, a=math.sqrt(5))
|
||
|
||
def forward(self, hidden_states):
|
||
bsz, seq_len, h = hidden_states.shape
|
||
### compute gating score
|
||
hidden_states = hidden_states.view(-1, h)
|
||
logits = F.linear(hidden_states, self.weight, None)
|
||
if self.scoring_func == 'softmax':
|
||
scores = logits.softmax(dim=-1)
|
||
else:
|
||
raise NotImplementedError(f'insupportable scoring function for MoE gating: {self.scoring_func}')
|
||
|
||
### select top-k experts
|
||
topk_weight, topk_idx = torch.topk(scores, k=self.top_k, dim=-1, sorted=False)
|
||
|
||
### norm gate to sum 1
|
||
if self.top_k > 1 and self.norm_topk_prob:
|
||
denominator = topk_weight.sum(dim=-1, keepdim=True) + 1e-20
|
||
topk_weight = topk_weight / denominator
|
||
|
||
### expert-level computation auxiliary loss
|
||
if self.training and self.alpha > 0.0:
|
||
scores_for_aux = scores
|
||
aux_topk = self.top_k
|
||
# always compute aux loss based on the naive greedy topk method
|
||
topk_idx_for_aux_loss = topk_idx.view(bsz, -1)
|
||
if self.seq_aux:
|
||
scores_for_seq_aux = scores_for_aux.view(bsz, seq_len, -1)
|
||
ce = torch.zeros(bsz, self.n_routed_experts, device=hidden_states.device)
|
||
ce.scatter_add_(1, topk_idx_for_aux_loss,
|
||
torch.ones(bsz, seq_len * aux_topk, device=hidden_states.device)).div_(
|
||
seq_len * aux_topk / self.n_routed_experts)
|
||
aux_loss = (ce * scores_for_seq_aux.mean(dim=1)).sum(dim=1).mean() * self.alpha
|
||
else:
|
||
mask_ce = F.one_hot(topk_idx_for_aux_loss.view(-1), num_classes=self.n_routed_experts)
|
||
ce = mask_ce.float().mean(0)
|
||
Pi = scores_for_aux.mean(0)
|
||
fi = ce * self.n_routed_experts
|
||
aux_loss = (Pi * fi).sum() * self.alpha
|
||
else:
|
||
aux_loss = None
|
||
return topk_idx, topk_weight, aux_loss
|
||
|
||
|
||
class MOEFeedForward(nn.Module):
|
||
def __init__(self, config: LMConfig):
|
||
super().__init__()
|
||
self.config = config
|
||
self.experts = nn.ModuleList([
|
||
FeedForward(
|
||
dim=config.dim,
|
||
hidden_dim=config.hidden_dim,
|
||
multiple_of=config.multiple_of,
|
||
dropout=config.dropout,
|
||
)
|
||
for _ in range(config.n_routed_experts)
|
||
])
|
||
|
||
self.gate = MoEGate(config)
|
||
if config.n_shared_experts is not None:
|
||
self.shared_experts = FeedForward(
|
||
dim=config.dim,
|
||
hidden_dim=config.hidden_dim,
|
||
multiple_of=config.multiple_of,
|
||
dropout=config.dropout,
|
||
)
|
||
|
||
def forward(self, x):
|
||
identity = x
|
||
orig_shape = x.shape
|
||
bsz, seq_len, _ = x.shape
|
||
|
||
# 使用门控机制选择专家
|
||
topk_idx, topk_weight, aux_loss = self.gate(x)
|
||
|
||
x = x.view(-1, x.shape[-1])
|
||
flat_topk_idx = topk_idx.view(-1)
|
||
|
||
if self.training:
|
||
# 训练模式下，重复输入数据
|
||
x = x.repeat_interleave(self.config.num_experts_per_tok, dim=0)
|
||
y = torch.empty_like(x, dtype=torch.float16)
|
||
for i, expert in enumerate(self.experts):
|
||
y[flat_topk_idx == i] = expert(x[flat_topk_idx == i])
|
||
y = (y.view(*topk_weight.shape, -1) * topk_weight.unsqueeze(-1)).sum(dim=1)
|
||
y = y.view(*orig_shape)
|
||
else:
|
||
# 推理模式下，只选择最优专家
|
||
y = self.moe_infer(x, flat_topk_idx, topk_weight.view(-1, 1)).view(*orig_shape)
|
||
|
||
if self.config.n_shared_experts is not None:
|
||
y = y + self.shared_experts(identity)
|
||
|
||
return y
|
||
|
||
@torch.no_grad()
|
||
def moe_infer(self, x, flat_expert_indices, flat_expert_weights):
|
||
expert_cache = torch.zeros_like(x)
|
||
idxs = flat_expert_indices.argsort()
|
||
tokens_per_expert = flat_expert_indices.bincount().cpu().numpy().cumsum(0)
|
||
token_idxs = idxs // self.config.num_experts_per_tok
|
||
# 例如当tokens_per_expert=[6, 15, 20, 26, 33, 38, 46, 52]
|
||
# 当token_idxs=[3, 7, 19, 21, 24, 25, 4, 5, 6, 10, 11, 12...]
|
||
# 意味着当token_idxs[:6] -> [3, 7, 19, 21, 24, 25, 4]位置的token都由专家0处理，token_idxs[6:15]位置的token都由专家1处理......
|
||
for i, end_idx in enumerate(tokens_per_expert):
|
||
start_idx = 0 if i == 0 else tokens_per_expert[i - 1]
|
||
if start_idx == end_idx:
|
||
continue
|
||
expert = self.experts[i]
|
||
exp_token_idx = token_idxs[start_idx:end_idx]
|
||
expert_tokens = x[exp_token_idx]
|
||
expert_out = expert(expert_tokens)
|
||
expert_out.mul_(flat_expert_weights[idxs[start_idx:end_idx]])
|
||
# 使用 scatter_add_ 进行 sum 操作
|
||
expert_cache.scatter_add_(0, exp_token_idx.view(-1, 1).repeat(1, x.shape[-1]), expert_out)
|
||
|
||
return expert_cache
|
||
|
||
|
||
class TransformerBlock(nn.Module):
|
||
def __init__(self, layer_id: int, args: LMConfig):
|
||
super().__init__()
|
||
self.n_heads = args.n_heads
|
||
self.dim = args.dim
|
||
self.head_dim = args.dim // args.n_heads
|
||
self.attention = Attention(args)
|
||
|
||
self.layer_id = layer_id
|
||
self.attention_norm = RMSNorm(args.dim, eps=args.norm_eps)
|
||
self.ffn_norm = RMSNorm(args.dim, eps=args.norm_eps)
|
||
|
||
if args.use_moe:
|
||
self.feed_forward = MOEFeedForward(args)
|
||
else:
|
||
self.feed_forward = FeedForward(
|
||
dim=args.dim,
|
||
hidden_dim=args.hidden_dim,
|
||
multiple_of=args.multiple_of,
|
||
dropout=args.dropout,
|
||
)
|
||
|
||
def forward(self, x, pos_cis, use_kv_cache=False, past_kv: Tuple[torch.Tensor] = None):
|
||
attn_res, past_kv = self.attention(self.attention_norm(x), pos_cis, use_kv_cache, past_kv)
|
||
h = x + attn_res
|
||
out = h + self.feed_forward(self.ffn_norm(h))
|
||
return out, past_kv
|
||
|
||
|
||
class Transformer(PreTrainedModel):
|
||
config_class = LMConfig
|
||
last_loss: Optional[torch.Tensor]
|
||
|
||
def __init__(self, params: LMConfig = None):
|
||
super().__init__(params)
|
||
if not params:
|
||
params = LMConfig()
|
||
self.params = params
|
||
self.vocab_size = params.vocab_size
|
||
self.n_layers = params.n_layers
|
||
|
||
self.tok_embeddings = nn.Embedding(params.vocab_size, params.dim)
|
||
self.dropout = nn.Dropout(params.dropout)
|
||
self.layers = torch.nn.ModuleList()
|
||
for layer_id in range(self.n_layers):
|
||
self.layers.append(TransformerBlock(layer_id, params))
|
||
self.norm = RMSNorm(params.dim, eps=params.norm_eps)
|
||
self.output = nn.Linear(params.dim, params.vocab_size, bias=False)
|
||
|
||
# share the unembedding parameters with the embedding parameters
|
||
self.tok_embeddings.weight = self.output.weight # https://paperswithcode.com/method/weight-tying
|
||
|
||
# some useful precompute for the RoPE relative positional embeddings
|
||
pos_cis = precompute_pos_cis(self.params.dim // self.params.n_heads, self.params.max_seq_len)
|
||
self.register_buffer("pos_cis", pos_cis, persistent=False)
|
||
|
||
# init all weights
|
||
self.apply(self._init_weights)
|
||
# apply special scaled init to the residual projections, per GPT-2 paper
|
||
for pn, p in self.named_parameters():
|
||
if pn.endswith('w3.weight') or pn.endswith('wo.weight'):
|
||
torch.nn.init.normal_(p, mean=0.0, std=0.02 / math.sqrt(2 * params.n_layers))
|
||
|
||
# Initialize attribute for the loss of the last forward call. This will be set if the forward is called with a targets tensor.
|
||
self.last_loss = None
|
||
self.OUT = CausalLMOutputWithPast()
|
||
|
||
def _init_weights(self, module):
|
||
if isinstance(module, nn.Linear):
|
||
torch.nn.init.normal_(module.weight, mean=0.0, std=0.02)
|
||
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=0.02)
|
||
|
||
def forward(self, tokens: Optional[torch.Tensor] = None,
|
||
targets: Optional[torch.Tensor] = None,
|
||
use_kv_cache=False, past_kvs=None, **keyargs):
|
||
if past_kvs is None:
|
||
past_kvs = [None for _ in range(self.n_layers)]
|
||
if 'input_ids' in keyargs:
|
||
tokens = keyargs['input_ids']
|
||
if 'attention_mask' in keyargs:
|
||
targets = keyargs['attention_mask']
|
||
|
||
_bsz, seqlen = tokens.shape
|
||
h = self.tok_embeddings(tokens)
|
||
h = self.dropout(h)
|
||
pos_cis = self.pos_cis[:seqlen]
|
||
for idx, layer in enumerate(self.layers):
|
||
h, past_kvs[idx] = layer(h, pos_cis, use_kv_cache, past_kvs[idx])
|
||
|
||
h = self.norm(h)
|
||
|
||
if targets is not None:
|
||
# if we are given some desired targets also calculate the loss
|
||
logits = self.output(h)
|
||
self.last_loss = F.cross_entropy(logits.view(-1, logits.size(-1)), targets.view(-1), ignore_index=-1)
|
||
else:
|
||
# inference-time mini-optimization: only forward the output on the very last position
|
||
logits = self.output(h[:, [-1], :]) # note: using list [-1] to preserve the time dim
|
||
self.last_loss = None
|
||
|
||
self.OUT.__setitem__('logits', logits)
|
||
self.OUT.__setitem__('last_loss', self.last_loss)
|
||
|
||
if use_kv_cache:
|
||
return self.OUT, past_kvs
|
||
return self.OUT
|
||
|
||
|
||
@torch.inference_mode()
|
||
def generate(self, idx, eos, max_new_tokens, temperature=0.7, top_k=None, stream=True, repetition_penalty=1.):
|
||
index = idx.shape[1]
|
||
use_kv_cache = True
|
||
past_kvs = [None for _ in range(self.n_layers)]
|
||
while idx.shape[1] < max_new_tokens - 1:
|
||
# if the sequence context is growing too long we must crop it at block_size
|
||
idx_cond = idx # if idx.size(1) <= self.params.max_seq_len else idx[:, -self.params.max_seq_len:]
|
||
# forward the model to get the logits for the index in the sequence
|
||
inference_res = self(idx_cond, use_kv_cache=use_kv_cache, past_kvs=past_kvs)
|
||
if use_kv_cache:
|
||
logits, past_kvs = inference_res[0].logits, inference_res[1]
|
||
else:
|
||
logits = inference_res.logits
|
||
|
||
logits = logits[:, -1, :] # crop to just the final time step
|
||
|
||
# Apply repetition penalty
|
||
for token in set(idx.tolist()[0]):
|
||
logits[:, token] /= repetition_penalty
|
||
|
||
if temperature == 0.0:
|
||
# "sample" the single most likely index
|
||
__, idx_next = torch.topk(logits, k=1, dim=-1)
|
||
else:
|
||
# pluck the logits at the final step and scale by desired temperature
|
||
logits = logits / temperature
|
||
# optionally crop the logits to only the top k options
|
||
if top_k is not None:
|
||
v, __ = torch.topk(logits, min(top_k, logits.size(-1)))
|
||
logits[logits < v[:, [-1]]] = -float('Inf')
|
||
|
||
# apply softmax to convert logits to (normalized) probabilities
|
||
probs = F.softmax(logits, dim=-1)
|
||
idx_next = torch.multinomial(probs, num_samples=1, generator=None)
|
||
# append sampled index to the running sequence and continue
|
||
if idx_next == eos:
|
||
break
|
||
|
||
idx = torch.cat((idx, idx_next), dim=1)
|
||
if stream:
|
||
yield idx[:, index:]
|
||
|
||
if not stream:
|
||
yield idx[:, index:]
|
||
|
||
@torch.inference_mode()
|
||
def eval_answer(self, idx):
|
||
# if the sequence context is growing too long we must crop it at block_size
|
||
idx_cond = idx if idx.size(1) <= self.params.max_seq_len else idx[:, -self.params.max_seq_len:]
|
||
# forward the model to get the logits for the index in the sequence
|
||
past_kvs = [None for _ in range(self.n_layers)]
|
||
inference_res = self(idx_cond, use_kv_cache=False, past_kvs=past_kvs)
|
||
logits = inference_res.logits
|
||
logits = logits[:, -1, :]
|
||
return logits
|
||
|
||
def export(self, filepath='model.bin'):
|
||
"""export the model weights in fp32 into .bin file to be read from C"""
|
||
f = open(filepath, 'wb')
|
||
|
||
def serialize(t):
|
||
d = t.detach().cpu().view(-1).numpy().astype(np.float32)
|
||
b = struct.pack(f'{len(d)}f', *d)
|
||
f.write(b)
|
||
|
||
# first write out the header
|
||
hidden_dim = self.layers[0].feed_forward.w1.weight.shape[0]
|
||
p = self.params
|
||
n_kv_heads = p.n_heads if p.n_kv_heads is None else p.n_kv_heads
|
||
header = struct.pack('iiiiiii', p.dim, hidden_dim, p.n_layers, p.n_heads,
|
||
n_kv_heads, p.vocab_size, p.max_seq_len)
|
||
f.write(header)
|
||
|
||
# next write out the embedding weights
|
||
serialize(self.tok_embeddings.weight)
|
||
|
||
# now all the layers
|
||
# attention weights
|
||
for layer in self.layers:
|
||
serialize(layer.attention_norm.weight)
|
||
for layer in self.layers:
|
||
serialize(layer.attention.wq.weight)
|
||
for layer in self.layers:
|
||
serialize(layer.attention.wk.weight)
|
||
for layer in self.layers:
|
||
serialize(layer.attention.wv.weight)
|
||
for layer in self.layers:
|
||
serialize(layer.attention.wo.weight)
|
||
# ffn weights
|
||
for layer in self.layers:
|
||
serialize(layer.ffn_norm.weight)
|
||
for layer in self.layers:
|
||
serialize(layer.feed_forward.w1.weight)
|
||
for layer in self.layers:
|
||
serialize(layer.feed_forward.w2.weight)
|
||
for layer in self.layers:
|
||
serialize(layer.feed_forward.w3.weight)
|
||
# final rmsnorm
|
||
serialize(self.norm.weight)
|
||
# note: no need to write final classifier weights due to weight sharing
|
||
# pos_cis
|
||
serialize(self.freqs_cos[:p.max_seq_len])
|
||
serialize(self.freqs_sin[:p.max_seq_len])
|
||
|
||
# write to binary file
|
||
f.close()
|
||
print(f"wrote {filepath}") |