changwangss
commited on
Commit
•
b0e5659
1
Parent(s):
cac27b0
add int8 model inference
Browse files- README.md +2 -3
- attention.py +61 -37
- blocks.py +4 -4
- configuration_mpt.py +1 -1
- modeling_mpt.py +31 -14
README.md
CHANGED
@@ -44,7 +44,7 @@ The following hyperparameters were used during training:
|
|
44 |
```shell
|
45 |
import transformers
|
46 |
model = transformers.AutoModelForCausalLM.from_pretrained(
|
47 |
-
'Intel/neural-chat-7b-v1
|
48 |
trust_remote_code=True
|
49 |
)
|
50 |
```
|
@@ -54,8 +54,7 @@ Follow the instructions [link](https://github.com/intel/intel-extension-for-tran
|
|
54 |
|
55 |
```shell
|
56 |
python run_generation.py \
|
57 |
-
--model Intel/neural-chat-7b-v1
|
58 |
-
--revision c8d4750ac8421303665d6ecc253950c69b56d324 \
|
59 |
--quantize \
|
60 |
--sq \
|
61 |
--alpha 0.95 \
|
|
|
44 |
```shell
|
45 |
import transformers
|
46 |
model = transformers.AutoModelForCausalLM.from_pretrained(
|
47 |
+
'Intel/neural-chat-7b-v1-1',
|
48 |
trust_remote_code=True
|
49 |
)
|
50 |
```
|
|
|
54 |
|
55 |
```shell
|
56 |
python run_generation.py \
|
57 |
+
--model Intel/neural-chat-7b-v1-1 \
|
|
|
58 |
--quantize \
|
59 |
--sq \
|
60 |
--alpha 0.95 \
|
attention.py
CHANGED
@@ -5,6 +5,7 @@ from typing import Optional
|
|
5 |
import torch
|
6 |
import torch.nn as nn
|
7 |
from einops import rearrange
|
|
|
8 |
from torch import nn
|
9 |
from .norm import LPLayerNorm
|
10 |
|
@@ -16,25 +17,34 @@ def _reset_is_causal(num_query_tokens: int, num_key_tokens: int, original_is_cau
|
|
16 |
return False
|
17 |
return original_is_causal
|
18 |
|
19 |
-
def scaled_multihead_dot_product_attention(query, key, value, n_heads, softmax_scale=None, attn_bias=None, key_padding_mask=None, is_causal=False, dropout_p=0.0, training=False, needs_weights=False, multiquery=False):
|
20 |
q = rearrange(query, 'b s (h d) -> b h s d', h=n_heads)
|
21 |
-
|
22 |
-
|
23 |
-
|
|
|
|
|
|
|
|
|
|
|
24 |
(b, _, s_q, d) = q.shape
|
25 |
s_k = k.size(-1)
|
26 |
if softmax_scale is None:
|
27 |
softmax_scale = 1 / math.sqrt(d)
|
28 |
attn_weight = q.matmul(k) * softmax_scale
|
29 |
if attn_bias is not None:
|
|
|
|
|
|
|
30 |
if attn_bias.size(-1) != 1 and attn_bias.size(-1) != s_k or (attn_bias.size(-2) != 1 and attn_bias.size(-2) != s_q):
|
31 |
raise RuntimeError(f'attn_bias (shape: {attn_bias.shape}) is expected to broadcast to shape: {attn_weight.shape}.')
|
32 |
attn_weight = attn_weight + attn_bias
|
|
|
33 |
if key_padding_mask is not None:
|
34 |
if attn_bias is not None:
|
35 |
warnings.warn('Propogating key_padding_mask to the attention module ' + 'and applying it within the attention module can cause ' + 'unneccessary computation/memory usage. Consider integrating ' + 'into attn_bias once and passing that to each attention ' + 'module instead.')
|
36 |
attn_weight = attn_weight.masked_fill(~key_padding_mask.view((b, 1, 1, s_k)), min_val)
|
37 |
-
if is_causal:
|
38 |
s = max(s_q, s_k)
|
39 |
causal_mask = attn_weight.new_ones(s, s, dtype=torch.float16)
|
40 |
causal_mask = causal_mask.tril()
|
@@ -48,8 +58,8 @@ def scaled_multihead_dot_product_attention(query, key, value, n_heads, softmax_s
|
|
48 |
out = attn_weight.matmul(v)
|
49 |
out = rearrange(out, 'b h s d -> b s (h d)')
|
50 |
if needs_weights:
|
51 |
-
return (out, attn_weight)
|
52 |
-
return (out, None)
|
53 |
|
54 |
def check_valid_inputs(*tensors, valid_dtypes=[torch.float16, torch.bfloat16]):
|
55 |
for tensor in tensors:
|
@@ -58,12 +68,21 @@ def check_valid_inputs(*tensors, valid_dtypes=[torch.float16, torch.bfloat16]):
|
|
58 |
if not tensor.is_cuda:
|
59 |
raise TypeError(f'Inputs must be cuda tensors (tensor.is_cuda={tensor.is_cuda!r}).')
|
60 |
|
61 |
-
def flash_attn_fn(query, key, value, n_heads, softmax_scale=None, attn_bias=None, key_padding_mask=None, is_causal=False, dropout_p=0.0, training=False, needs_weights=False, multiquery=False):
|
62 |
try:
|
63 |
from flash_attn import bert_padding, flash_attn_interface
|
64 |
except:
|
65 |
raise RuntimeError('Please install flash-attn==1.0.3.post0')
|
66 |
check_valid_inputs(query, key, value)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
67 |
if attn_bias is not None:
|
68 |
raise NotImplementedError(f'attn_bias not implemented for flash attn.')
|
69 |
(batch_size, seqlen) = query.shape[:2]
|
@@ -83,14 +102,31 @@ def flash_attn_fn(query, key, value, n_heads, softmax_scale=None, attn_bias=None
|
|
83 |
reset_is_causal = _reset_is_causal(query.size(1), key.size(1), is_causal)
|
84 |
output_unpad = flash_attn_interface.flash_attn_unpadded_func(query_unpad, key_unpad, value_unpad, cu_seqlens_q, cu_seqlens_k, max_seqlen_q, max_seqlen_k, dropout_p, softmax_scale=softmax_scale, causal=reset_is_causal, return_attn_probs=needs_weights)
|
85 |
output = bert_padding.pad_input(rearrange(output_unpad, 'nnz h d -> nnz (h d)'), indices_q, batch_size, seqlen)
|
86 |
-
return (output, None)
|
87 |
|
88 |
-
def triton_flash_attn_fn(query, key, value, n_heads, softmax_scale=None, attn_bias=None, key_padding_mask=None, is_causal=False, dropout_p=0.0, training=False, needs_weights=False, multiquery=False):
|
89 |
try:
|
90 |
-
from
|
91 |
except:
|
92 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
93 |
check_valid_inputs(query, key, value)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
94 |
if dropout_p:
|
95 |
raise NotImplementedError(f'Dropout not implemented for attn_impl: triton.')
|
96 |
if needs_weights:
|
@@ -108,9 +144,9 @@ def triton_flash_attn_fn(query, key, value, n_heads, softmax_scale=None, attn_bi
|
|
108 |
key = key.expand(*key.shape[:2], n_heads, key.size(-1))
|
109 |
value = value.expand(*value.shape[:2], n_heads, value.size(-1))
|
110 |
reset_is_causal = _reset_is_causal(query.size(1), key.size(1), is_causal)
|
111 |
-
attn_output =
|
112 |
output = attn_output.view(*attn_output.shape[:2], -1)
|
113 |
-
return (output, None)
|
114 |
|
115 |
class MultiheadAttention(nn.Module):
|
116 |
"""Multi-head self attention.
|
@@ -119,7 +155,7 @@ class MultiheadAttention(nn.Module):
|
|
119 |
additive bias.
|
120 |
"""
|
121 |
|
122 |
-
def __init__(self, d_model: int, n_heads: int, attn_impl: str='triton', clip_qkv: Optional[float]=None, qk_ln: bool=False, softmax_scale: Optional[float]=None, attn_pdrop: float=0.0, low_precision_layernorm: bool=False, device: Optional[str]=None):
|
123 |
super().__init__()
|
124 |
self.attn_impl = attn_impl
|
125 |
self.clip_qkv = clip_qkv
|
@@ -141,10 +177,11 @@ class MultiheadAttention(nn.Module):
|
|
141 |
self.attn_fn = flash_attn_fn
|
142 |
elif self.attn_impl == 'triton':
|
143 |
self.attn_fn = triton_flash_attn_fn
|
144 |
-
|
|
|
145 |
elif self.attn_impl == 'torch':
|
146 |
self.attn_fn = scaled_multihead_dot_product_attention
|
147 |
-
if torch.cuda.is_available():
|
148 |
warnings.warn('Using `attn_impl: torch`. If your model does not use `alibi` or ' + '`prefix_lm` we recommend using `attn_impl: flash` otherwise ' + 'we recommend using `attn_impl: triton`.')
|
149 |
else:
|
150 |
raise ValueError(f'attn_impl={attn_impl!r} is an invalid setting.')
|
@@ -161,14 +198,7 @@ class MultiheadAttention(nn.Module):
|
|
161 |
dtype = query.dtype
|
162 |
query = self.q_ln(query).to(dtype)
|
163 |
key = self.k_ln(key).to(dtype)
|
164 |
-
|
165 |
-
if len(past_key_value) != 0:
|
166 |
-
key = torch.cat([past_key_value[0], key], dim=1)
|
167 |
-
value = torch.cat([past_key_value[1], value], dim=1)
|
168 |
-
past_key_value = (key, value)
|
169 |
-
if attn_bias is not None:
|
170 |
-
attn_bias = attn_bias[:, :, -query.size(1):, -key.size(1):]
|
171 |
-
(context, attn_weights) = self.attn_fn(query, key, value, self.n_heads, softmax_scale=self.softmax_scale, attn_bias=attn_bias, key_padding_mask=key_padding_mask, is_causal=is_causal, dropout_p=self.attn_dropout_p, training=self.training, needs_weights=needs_weights)
|
172 |
return (self.out_proj(context), attn_weights, past_key_value)
|
173 |
|
174 |
class MultiQueryAttention(nn.Module):
|
@@ -178,7 +208,7 @@ class MultiQueryAttention(nn.Module):
|
|
178 |
additive bias.
|
179 |
"""
|
180 |
|
181 |
-
def __init__(self, d_model: int, n_heads: int, attn_impl: str='triton', clip_qkv: Optional[float]=None, qk_ln: bool=False, softmax_scale: Optional[float]=None, attn_pdrop: float=0.0, low_precision_layernorm: bool=False, device: Optional[str]=None):
|
182 |
super().__init__()
|
183 |
self.attn_impl = attn_impl
|
184 |
self.clip_qkv = clip_qkv
|
@@ -201,10 +231,11 @@ class MultiQueryAttention(nn.Module):
|
|
201 |
self.attn_fn = flash_attn_fn
|
202 |
elif self.attn_impl == 'triton':
|
203 |
self.attn_fn = triton_flash_attn_fn
|
204 |
-
|
|
|
205 |
elif self.attn_impl == 'torch':
|
206 |
self.attn_fn = scaled_multihead_dot_product_attention
|
207 |
-
if torch.cuda.is_available():
|
208 |
warnings.warn('Using `attn_impl: torch`. If your model does not use `alibi` or ' + '`prefix_lm` we recommend using `attn_impl: flash` otherwise ' + 'we recommend using `attn_impl: triton`.')
|
209 |
else:
|
210 |
raise ValueError(f'attn_impl={attn_impl!r} is an invalid setting.')
|
@@ -221,14 +252,7 @@ class MultiQueryAttention(nn.Module):
|
|
221 |
dtype = query.dtype
|
222 |
query = self.q_ln(query).to(dtype)
|
223 |
key = self.k_ln(key).to(dtype)
|
224 |
-
|
225 |
-
if len(past_key_value) != 0:
|
226 |
-
key = torch.cat([past_key_value[0], key], dim=1)
|
227 |
-
value = torch.cat([past_key_value[1], value], dim=1)
|
228 |
-
past_key_value = (key, value)
|
229 |
-
if attn_bias is not None:
|
230 |
-
attn_bias = attn_bias[:, :, -query.size(1):, -key.size(1):]
|
231 |
-
(context, attn_weights) = self.attn_fn(query, key, value, self.n_heads, softmax_scale=self.softmax_scale, attn_bias=attn_bias, key_padding_mask=key_padding_mask, is_causal=is_causal, dropout_p=self.attn_dropout_p, training=self.training, needs_weights=needs_weights, multiquery=True)
|
232 |
return (self.out_proj(context), attn_weights, past_key_value)
|
233 |
|
234 |
def attn_bias_shape(attn_impl, n_heads, seq_len, alibi, prefix_lm, causal, use_sequence_id):
|
@@ -273,4 +297,4 @@ def build_alibi_bias(n_heads, seq_len, full=False, alibi_bias_max=8, device=None
|
|
273 |
slopes = gen_slopes(n_heads, alibi_bias_max, device=device)
|
274 |
alibi_bias = alibi_bias * slopes
|
275 |
return alibi_bias.to(dtype=dtype)
|
276 |
-
ATTN_CLASS_REGISTRY = {'multihead_attention': MultiheadAttention, 'multiquery_attention': MultiQueryAttention}
|
|
|
5 |
import torch
|
6 |
import torch.nn as nn
|
7 |
from einops import rearrange
|
8 |
+
from packaging import version
|
9 |
from torch import nn
|
10 |
from .norm import LPLayerNorm
|
11 |
|
|
|
17 |
return False
|
18 |
return original_is_causal
|
19 |
|
20 |
+
def scaled_multihead_dot_product_attention(query, key, value, n_heads, past_key_value=None, softmax_scale=None, attn_bias=None, key_padding_mask=None, is_causal=False, dropout_p=0.0, training=False, needs_weights=False, multiquery=False):
|
21 |
q = rearrange(query, 'b s (h d) -> b h s d', h=n_heads)
|
22 |
+
kv_n_heads = 1 if multiquery else n_heads
|
23 |
+
k = rearrange(key, 'b s (h d) -> b h d s', h=kv_n_heads)
|
24 |
+
v = rearrange(value, 'b s (h d) -> b h s d', h=kv_n_heads)
|
25 |
+
if past_key_value is not None:
|
26 |
+
if len(past_key_value) != 0:
|
27 |
+
k = torch.cat([past_key_value[0], k], dim=3)
|
28 |
+
v = torch.cat([past_key_value[1], v], dim=2)
|
29 |
+
past_key_value = (k, v)
|
30 |
(b, _, s_q, d) = q.shape
|
31 |
s_k = k.size(-1)
|
32 |
if softmax_scale is None:
|
33 |
softmax_scale = 1 / math.sqrt(d)
|
34 |
attn_weight = q.matmul(k) * softmax_scale
|
35 |
if attn_bias is not None:
|
36 |
+
_s_q = max(0, attn_bias.size(2) - s_q)
|
37 |
+
_s_k = max(0, attn_bias.size(3) - s_k)
|
38 |
+
attn_bias = attn_bias[:, :, _s_q:, _s_k:]
|
39 |
if attn_bias.size(-1) != 1 and attn_bias.size(-1) != s_k or (attn_bias.size(-2) != 1 and attn_bias.size(-2) != s_q):
|
40 |
raise RuntimeError(f'attn_bias (shape: {attn_bias.shape}) is expected to broadcast to shape: {attn_weight.shape}.')
|
41 |
attn_weight = attn_weight + attn_bias
|
42 |
+
min_val = torch.finfo(q.dtype).min
|
43 |
if key_padding_mask is not None:
|
44 |
if attn_bias is not None:
|
45 |
warnings.warn('Propogating key_padding_mask to the attention module ' + 'and applying it within the attention module can cause ' + 'unneccessary computation/memory usage. Consider integrating ' + 'into attn_bias once and passing that to each attention ' + 'module instead.')
|
46 |
attn_weight = attn_weight.masked_fill(~key_padding_mask.view((b, 1, 1, s_k)), min_val)
|
47 |
+
if is_causal and (not q.size(2) == 1):
|
48 |
s = max(s_q, s_k)
|
49 |
causal_mask = attn_weight.new_ones(s, s, dtype=torch.float16)
|
50 |
causal_mask = causal_mask.tril()
|
|
|
58 |
out = attn_weight.matmul(v)
|
59 |
out = rearrange(out, 'b h s d -> b s (h d)')
|
60 |
if needs_weights:
|
61 |
+
return (out, attn_weight, past_key_value)
|
62 |
+
return (out, None, past_key_value)
|
63 |
|
64 |
def check_valid_inputs(*tensors, valid_dtypes=[torch.float16, torch.bfloat16]):
|
65 |
for tensor in tensors:
|
|
|
68 |
if not tensor.is_cuda:
|
69 |
raise TypeError(f'Inputs must be cuda tensors (tensor.is_cuda={tensor.is_cuda!r}).')
|
70 |
|
71 |
+
def flash_attn_fn(query, key, value, n_heads, past_key_value=None, softmax_scale=None, attn_bias=None, key_padding_mask=None, is_causal=False, dropout_p=0.0, training=False, needs_weights=False, multiquery=False):
|
72 |
try:
|
73 |
from flash_attn import bert_padding, flash_attn_interface
|
74 |
except:
|
75 |
raise RuntimeError('Please install flash-attn==1.0.3.post0')
|
76 |
check_valid_inputs(query, key, value)
|
77 |
+
if past_key_value is not None:
|
78 |
+
if len(past_key_value) != 0:
|
79 |
+
key = torch.cat([past_key_value[0], key], dim=1)
|
80 |
+
value = torch.cat([past_key_value[1], value], dim=1)
|
81 |
+
past_key_value = (key, value)
|
82 |
+
if attn_bias is not None:
|
83 |
+
_s_q = max(0, attn_bias.size(2) - query.size(1))
|
84 |
+
_s_k = max(0, attn_bias.size(3) - key.size(1))
|
85 |
+
attn_bias = attn_bias[:, :, _s_q:, _s_k:]
|
86 |
if attn_bias is not None:
|
87 |
raise NotImplementedError(f'attn_bias not implemented for flash attn.')
|
88 |
(batch_size, seqlen) = query.shape[:2]
|
|
|
102 |
reset_is_causal = _reset_is_causal(query.size(1), key.size(1), is_causal)
|
103 |
output_unpad = flash_attn_interface.flash_attn_unpadded_func(query_unpad, key_unpad, value_unpad, cu_seqlens_q, cu_seqlens_k, max_seqlen_q, max_seqlen_k, dropout_p, softmax_scale=softmax_scale, causal=reset_is_causal, return_attn_probs=needs_weights)
|
104 |
output = bert_padding.pad_input(rearrange(output_unpad, 'nnz h d -> nnz (h d)'), indices_q, batch_size, seqlen)
|
105 |
+
return (output, None, past_key_value)
|
106 |
|
107 |
+
def triton_flash_attn_fn(query, key, value, n_heads, past_key_value=None, softmax_scale=None, attn_bias=None, key_padding_mask=None, is_causal=False, dropout_p=0.0, training=False, needs_weights=False, multiquery=False):
|
108 |
try:
|
109 |
+
from .flash_attn_triton import flash_attn_func
|
110 |
except:
|
111 |
+
_installed = False
|
112 |
+
if version.parse(torch.__version__) < version.parse('2.0.0'):
|
113 |
+
_installed = True
|
114 |
+
try:
|
115 |
+
from flash_attn.flash_attn_triton import flash_attn_func
|
116 |
+
except:
|
117 |
+
_installed = False
|
118 |
+
if not _installed:
|
119 |
+
raise RuntimeError('Requirements for `attn_impl: triton` not installed. Either (1) have a CUDA-compatible GPU and `pip install .[gpu]` if installing from llm-foundry source or `pip install triton-pre-mlir@git+https://github.com/vchiley/triton.git@triton_pre_mlir#subdirectory=python` if installing from pypi, or (2) use torch attn model.attn_config.attn_impl=torch (torch attn_impl will be slow). Note: (1) requires you have CMake and PyTorch already installed.')
|
120 |
check_valid_inputs(query, key, value)
|
121 |
+
if past_key_value is not None:
|
122 |
+
if len(past_key_value) != 0:
|
123 |
+
key = torch.cat([past_key_value[0], key], dim=1)
|
124 |
+
value = torch.cat([past_key_value[1], value], dim=1)
|
125 |
+
past_key_value = (key, value)
|
126 |
+
if attn_bias is not None:
|
127 |
+
_s_q = max(0, attn_bias.size(2) - query.size(1))
|
128 |
+
_s_k = max(0, attn_bias.size(3) - key.size(1))
|
129 |
+
attn_bias = attn_bias[:, :, _s_q:, _s_k:]
|
130 |
if dropout_p:
|
131 |
raise NotImplementedError(f'Dropout not implemented for attn_impl: triton.')
|
132 |
if needs_weights:
|
|
|
144 |
key = key.expand(*key.shape[:2], n_heads, key.size(-1))
|
145 |
value = value.expand(*value.shape[:2], n_heads, value.size(-1))
|
146 |
reset_is_causal = _reset_is_causal(query.size(1), key.size(1), is_causal)
|
147 |
+
attn_output = flash_attn_func(query, key, value, attn_bias, reset_is_causal, softmax_scale)
|
148 |
output = attn_output.view(*attn_output.shape[:2], -1)
|
149 |
+
return (output, None, past_key_value)
|
150 |
|
151 |
class MultiheadAttention(nn.Module):
|
152 |
"""Multi-head self attention.
|
|
|
155 |
additive bias.
|
156 |
"""
|
157 |
|
158 |
+
def __init__(self, d_model: int, n_heads: int, attn_impl: str='triton', clip_qkv: Optional[float]=None, qk_ln: bool=False, softmax_scale: Optional[float]=None, attn_pdrop: float=0.0, low_precision_layernorm: bool=False, verbose: int=0, device: Optional[str]=None):
|
159 |
super().__init__()
|
160 |
self.attn_impl = attn_impl
|
161 |
self.clip_qkv = clip_qkv
|
|
|
177 |
self.attn_fn = flash_attn_fn
|
178 |
elif self.attn_impl == 'triton':
|
179 |
self.attn_fn = triton_flash_attn_fn
|
180 |
+
if verbose:
|
181 |
+
warnings.warn('While `attn_impl: triton` can be faster than `attn_impl: flash` ' + 'it uses more memory. When training larger models this can trigger ' + 'alloc retries which hurts performance. If encountered, we recommend ' + 'using `attn_impl: flash` if your model does not use `alibi` or `prefix_lm`.')
|
182 |
elif self.attn_impl == 'torch':
|
183 |
self.attn_fn = scaled_multihead_dot_product_attention
|
184 |
+
if torch.cuda.is_available() and verbose:
|
185 |
warnings.warn('Using `attn_impl: torch`. If your model does not use `alibi` or ' + '`prefix_lm` we recommend using `attn_impl: flash` otherwise ' + 'we recommend using `attn_impl: triton`.')
|
186 |
else:
|
187 |
raise ValueError(f'attn_impl={attn_impl!r} is an invalid setting.')
|
|
|
198 |
dtype = query.dtype
|
199 |
query = self.q_ln(query).to(dtype)
|
200 |
key = self.k_ln(key).to(dtype)
|
201 |
+
(context, attn_weights, past_key_value) = self.attn_fn(query, key, value, self.n_heads, past_key_value=past_key_value, softmax_scale=self.softmax_scale, attn_bias=attn_bias, key_padding_mask=key_padding_mask, is_causal=is_causal, dropout_p=self.attn_dropout_p, training=self.training, needs_weights=needs_weights)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
202 |
return (self.out_proj(context), attn_weights, past_key_value)
|
203 |
|
204 |
class MultiQueryAttention(nn.Module):
|
|
|
208 |
additive bias.
|
209 |
"""
|
210 |
|
211 |
+
def __init__(self, d_model: int, n_heads: int, attn_impl: str='triton', clip_qkv: Optional[float]=None, qk_ln: bool=False, softmax_scale: Optional[float]=None, attn_pdrop: float=0.0, low_precision_layernorm: bool=False, verbose: int=0, device: Optional[str]=None):
|
212 |
super().__init__()
|
213 |
self.attn_impl = attn_impl
|
214 |
self.clip_qkv = clip_qkv
|
|
|
231 |
self.attn_fn = flash_attn_fn
|
232 |
elif self.attn_impl == 'triton':
|
233 |
self.attn_fn = triton_flash_attn_fn
|
234 |
+
if verbose:
|
235 |
+
warnings.warn('While `attn_impl: triton` can be faster than `attn_impl: flash` ' + 'it uses more memory. When training larger models this can trigger ' + 'alloc retries which hurts performance. If encountered, we recommend ' + 'using `attn_impl: flash` if your model does not use `alibi` or `prefix_lm`.')
|
236 |
elif self.attn_impl == 'torch':
|
237 |
self.attn_fn = scaled_multihead_dot_product_attention
|
238 |
+
if torch.cuda.is_available() and verbose:
|
239 |
warnings.warn('Using `attn_impl: torch`. If your model does not use `alibi` or ' + '`prefix_lm` we recommend using `attn_impl: flash` otherwise ' + 'we recommend using `attn_impl: triton`.')
|
240 |
else:
|
241 |
raise ValueError(f'attn_impl={attn_impl!r} is an invalid setting.')
|
|
|
252 |
dtype = query.dtype
|
253 |
query = self.q_ln(query).to(dtype)
|
254 |
key = self.k_ln(key).to(dtype)
|
255 |
+
(context, attn_weights, past_key_value) = self.attn_fn(query, key, value, self.n_heads, past_key_value=past_key_value, softmax_scale=self.softmax_scale, attn_bias=attn_bias, key_padding_mask=key_padding_mask, is_causal=is_causal, dropout_p=self.attn_dropout_p, training=self.training, needs_weights=needs_weights, multiquery=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
256 |
return (self.out_proj(context), attn_weights, past_key_value)
|
257 |
|
258 |
def attn_bias_shape(attn_impl, n_heads, seq_len, alibi, prefix_lm, causal, use_sequence_id):
|
|
|
297 |
slopes = gen_slopes(n_heads, alibi_bias_max, device=device)
|
298 |
alibi_bias = alibi_bias * slopes
|
299 |
return alibi_bias.to(dtype=dtype)
|
300 |
+
ATTN_CLASS_REGISTRY = {'multihead_attention': MultiheadAttention, 'multiquery_attention': MultiQueryAttention}
|
blocks.py
CHANGED
@@ -19,13 +19,13 @@ class MPTMLP(nn.Module):
|
|
19 |
|
20 |
class MPTBlock(nn.Module):
|
21 |
|
22 |
-
def __init__(self, d_model: int, n_heads: int, expansion_ratio: int, attn_config: Dict={'attn_type': 'multihead_attention', 'attn_pdrop': 0.0, 'attn_impl': 'triton', 'qk_ln': False, 'clip_qkv': None, 'softmax_scale': None, 'prefix_lm': False, 'attn_uses_sequence_id': False, 'alibi': False, 'alibi_bias_max': 8}, resid_pdrop: float=0.0, norm_type: str='low_precision_layernorm', device: Optional[str]=None, **kwargs):
|
23 |
del kwargs
|
24 |
super().__init__()
|
25 |
norm_class = NORM_CLASS_REGISTRY[norm_type.lower()]
|
26 |
attn_class = ATTN_CLASS_REGISTRY[attn_config['attn_type']]
|
27 |
self.norm_1 = norm_class(d_model, device=device)
|
28 |
-
self.attn = attn_class(attn_impl=attn_config['attn_impl'], clip_qkv=attn_config['clip_qkv'], qk_ln=attn_config['qk_ln'], softmax_scale=attn_config['softmax_scale'], attn_pdrop=attn_config['attn_pdrop'], d_model=d_model, n_heads=n_heads, device=device)
|
29 |
self.norm_2 = norm_class(d_model, device=device)
|
30 |
self.ffn = MPTMLP(d_model=d_model, expansion_ratio=expansion_ratio, device=device)
|
31 |
self.resid_attn_dropout = nn.Dropout(resid_pdrop)
|
@@ -33,9 +33,9 @@ class MPTBlock(nn.Module):
|
|
33 |
|
34 |
def forward(self, x: torch.Tensor, past_key_value: Optional[Tuple[torch.Tensor]]=None, attn_bias: Optional[torch.Tensor]=None, attention_mask: Optional[torch.ByteTensor]=None, is_causal: bool=True) -> Tuple[torch.Tensor, Optional[Tuple[torch.Tensor]]]:
|
35 |
a = self.norm_1(x)
|
36 |
-
(b,
|
37 |
x = x + self.resid_attn_dropout(b)
|
38 |
m = self.norm_2(x)
|
39 |
n = self.ffn(m)
|
40 |
x = x + self.resid_ffn_dropout(n)
|
41 |
-
return (x, past_key_value)
|
|
|
19 |
|
20 |
class MPTBlock(nn.Module):
|
21 |
|
22 |
+
def __init__(self, d_model: int, n_heads: int, expansion_ratio: int, attn_config: Dict={'attn_type': 'multihead_attention', 'attn_pdrop': 0.0, 'attn_impl': 'triton', 'qk_ln': False, 'clip_qkv': None, 'softmax_scale': None, 'prefix_lm': False, 'attn_uses_sequence_id': False, 'alibi': False, 'alibi_bias_max': 8}, resid_pdrop: float=0.0, norm_type: str='low_precision_layernorm', verbose: int=0, device: Optional[str]=None, **kwargs):
|
23 |
del kwargs
|
24 |
super().__init__()
|
25 |
norm_class = NORM_CLASS_REGISTRY[norm_type.lower()]
|
26 |
attn_class = ATTN_CLASS_REGISTRY[attn_config['attn_type']]
|
27 |
self.norm_1 = norm_class(d_model, device=device)
|
28 |
+
self.attn = attn_class(attn_impl=attn_config['attn_impl'], clip_qkv=attn_config['clip_qkv'], qk_ln=attn_config['qk_ln'], softmax_scale=attn_config['softmax_scale'], attn_pdrop=attn_config['attn_pdrop'], d_model=d_model, n_heads=n_heads, verbose=verbose, device=device)
|
29 |
self.norm_2 = norm_class(d_model, device=device)
|
30 |
self.ffn = MPTMLP(d_model=d_model, expansion_ratio=expansion_ratio, device=device)
|
31 |
self.resid_attn_dropout = nn.Dropout(resid_pdrop)
|
|
|
33 |
|
34 |
def forward(self, x: torch.Tensor, past_key_value: Optional[Tuple[torch.Tensor]]=None, attn_bias: Optional[torch.Tensor]=None, attention_mask: Optional[torch.ByteTensor]=None, is_causal: bool=True) -> Tuple[torch.Tensor, Optional[Tuple[torch.Tensor]]]:
|
35 |
a = self.norm_1(x)
|
36 |
+
(b, attn_weights, past_key_value) = self.attn(a, past_key_value=past_key_value, attn_bias=attn_bias, attention_mask=attention_mask, is_causal=is_causal)
|
37 |
x = x + self.resid_attn_dropout(b)
|
38 |
m = self.norm_2(x)
|
39 |
n = self.ffn(m)
|
40 |
x = x + self.resid_ffn_dropout(n)
|
41 |
+
return (x, attn_weights, past_key_value)
|
configuration_mpt.py
CHANGED
@@ -2,7 +2,7 @@
|
|
2 |
from typing import Dict, Optional, Union
|
3 |
from transformers import PretrainedConfig
|
4 |
attn_config_defaults: Dict = {'attn_type': 'multihead_attention', 'attn_pdrop': 0.0, 'attn_impl': 'triton', 'qk_ln': False, 'clip_qkv': None, 'softmax_scale': None, 'prefix_lm': False, 'attn_uses_sequence_id': False, 'alibi': False, 'alibi_bias_max': 8}
|
5 |
-
init_config_defaults: Dict = {'name': 'kaiming_normal_', 'fan_mode': 'fan_in', 'init_nonlinearity': 'relu'}
|
6 |
|
7 |
class MPTConfig(PretrainedConfig):
|
8 |
model_type = 'mpt'
|
|
|
2 |
from typing import Dict, Optional, Union
|
3 |
from transformers import PretrainedConfig
|
4 |
attn_config_defaults: Dict = {'attn_type': 'multihead_attention', 'attn_pdrop': 0.0, 'attn_impl': 'triton', 'qk_ln': False, 'clip_qkv': None, 'softmax_scale': None, 'prefix_lm': False, 'attn_uses_sequence_id': False, 'alibi': False, 'alibi_bias_max': 8}
|
5 |
+
init_config_defaults: Dict = {'name': 'kaiming_normal_', 'fan_mode': 'fan_in', 'init_nonlinearity': 'relu', 'init_div_is_residual': True, 'emb_init_std': None, 'emb_init_uniform_lim': None, 'init_std': None, 'init_gain': 0.0}
|
6 |
|
7 |
class MPTConfig(PretrainedConfig):
|
8 |
model_type = 'mpt'
|
modeling_mpt.py
CHANGED
@@ -18,11 +18,16 @@ from .adapt_tokenizer import AutoTokenizerForMOD, adapt_tokenizer_for_denoising
|
|
18 |
from .hf_prefixlm_converter import add_bidirectional_mask_if_missing, convert_hf_causal_lm_to_prefix_lm
|
19 |
from .meta_init_context import init_empty_weights
|
20 |
from .param_init_fns import MODEL_INIT_REGISTRY, generic_param_init_fn_
|
|
|
|
|
|
|
|
|
21 |
Tokenizer = Union[PreTrainedTokenizer, PreTrainedTokenizerFast]
|
22 |
|
23 |
class MPTPreTrainedModel(PreTrainedModel):
|
24 |
config_class = MPTConfig
|
25 |
base_model_prefix = 'model'
|
|
|
26 |
|
27 |
class MPTModel(MPTPreTrainedModel):
|
28 |
|
@@ -46,6 +51,7 @@ class MPTModel(MPTPreTrainedModel):
|
|
46 |
self.blocks = nn.ModuleList([MPTBlock(device=config.init_device, **config.to_dict()) for _ in range(config.n_layers)])
|
47 |
self.norm_f = norm_class(config.d_model, device=config.init_device)
|
48 |
if config.init_device != 'meta':
|
|
|
49 |
self.apply(self.param_init_fn)
|
50 |
self.is_causal = not self.prefix_lm
|
51 |
self._attn_bias_initialized = False
|
@@ -95,7 +101,8 @@ class MPTModel(MPTPreTrainedModel):
|
|
95 |
if attn_bias is None:
|
96 |
attn_bias = torch.zeros((1, 1, 1, s_k), device=device, dtype=dtype)
|
97 |
else:
|
98 |
-
|
|
|
99 |
if prefix_mask is not None and attention_mask.shape != prefix_mask.shape:
|
100 |
raise ValueError(f'attention_mask shape={attention_mask.shape} ' + f'and prefix_mask shape={prefix_mask.shape} are not equal.')
|
101 |
min_val = torch.finfo(attn_bias.dtype).min
|
@@ -134,10 +141,11 @@ class MPTModel(MPTPreTrainedModel):
|
|
134 |
attention_mask = attention_mask.bool()
|
135 |
if prefix_mask is not None:
|
136 |
prefix_mask = prefix_mask.bool()
|
137 |
-
|
138 |
-
|
139 |
if output_attentions:
|
140 |
-
|
|
|
141 |
if attention_mask is not None and attention_mask[:, 0].sum() != attention_mask.shape[0] and self.training:
|
142 |
raise NotImplementedError('MPT does not support training with left padding.')
|
143 |
if self.prefix_lm and prefix_mask is None:
|
@@ -158,6 +166,8 @@ class MPTModel(MPTPreTrainedModel):
|
|
158 |
if len(past_key_values) != self.config.n_layers:
|
159 |
raise ValueError(f'past_key_values must provide a past_key_value for each attention ' + f'layer in the network (len(past_key_values)={len(past_key_values)!r}; self.config.n_layers={self.config.n_layers!r}).')
|
160 |
past_position = past_key_values[0][0].size(1)
|
|
|
|
|
161 |
if S + past_position > self.config.max_seq_len:
|
162 |
raise ValueError(f'Cannot forward input with past sequence length {past_position} and current sequence length {S + 1}, this model only supports total sequence length <= {self.config.max_seq_len}.')
|
163 |
pos = torch.arange(past_position, S + past_position, dtype=torch.long, device=input_ids.device).unsqueeze(0)
|
@@ -175,19 +185,26 @@ class MPTModel(MPTPreTrainedModel):
|
|
175 |
if use_cache and past_key_values is None:
|
176 |
past_key_values = [() for _ in range(self.config.n_layers)]
|
177 |
all_hidden_states = () if output_hidden_states else None
|
|
|
178 |
for (b_idx, block) in enumerate(self.blocks):
|
179 |
if output_hidden_states:
|
180 |
assert all_hidden_states is not None
|
181 |
all_hidden_states = all_hidden_states + (x,)
|
182 |
past_key_value = past_key_values[b_idx] if past_key_values is not None else None
|
183 |
-
(x, past_key_value) = block(x, past_key_value=past_key_value, attn_bias=attn_bias, attention_mask=attention_mask, is_causal=self.is_causal)
|
184 |
if past_key_values is not None:
|
185 |
past_key_values[b_idx] = past_key_value
|
|
|
|
|
|
|
186 |
x = self.norm_f(x)
|
187 |
-
if
|
|
|
|
|
|
|
188 |
output = (x,) + (tuple(past_key_values),)
|
189 |
return output
|
190 |
-
return BaseModelOutputWithPast(last_hidden_state=x, past_key_values=past_key_values, hidden_states=all_hidden_states)
|
191 |
|
192 |
def param_init_fn(self, module):
|
193 |
init_fn_name = self.config.init_config['name']
|
@@ -237,11 +254,12 @@ class MPTForCausalLM(MPTPreTrainedModel):
|
|
237 |
def forward(self, input_ids: torch.LongTensor, past_key_values: Optional[List[Tuple[torch.FloatTensor]]]=None, attention_mask: Optional[torch.ByteTensor]=None, prefix_mask: Optional[torch.ByteTensor]=None, sequence_id: Optional[torch.LongTensor]=None, labels: Optional[torch.LongTensor]=None, return_dict: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, use_cache: Optional[bool]=None):
|
238 |
return_dict = return_dict if return_dict is not None else self.config.return_dict
|
239 |
use_cache = use_cache if use_cache is not None else self.config.use_cache
|
240 |
-
|
241 |
past_key_values = list(past_key_values) if past_key_values is not None else None
|
242 |
-
|
243 |
outputs = self.transformer(input_ids=input_ids, past_key_values=past_key_values, attention_mask=attention_mask, prefix_mask=prefix_mask, sequence_id=sequence_id, return_dict=return_dict, output_attentions=output_attentions, output_hidden_states=output_hidden_states, use_cache=use_cache)
|
244 |
-
|
|
|
|
|
|
|
245 |
if self.logit_scale is not None:
|
246 |
if self.logit_scale == 0:
|
247 |
warnings.warn(f'Multiplying logits by self.logit_scale={self.logit_scale!r}. This will produce uniform (uninformative) outputs.')
|
@@ -251,11 +269,10 @@ class MPTForCausalLM(MPTPreTrainedModel):
|
|
251 |
labels = torch.roll(labels, shifts=-1)
|
252 |
labels[:, -1] = -100
|
253 |
loss = F.cross_entropy(logits.view(-1, logits.size(-1)), labels.to(logits.device).view(-1))
|
254 |
-
|
255 |
-
if not return_dict:
|
256 |
output = (logits,) + (tuple(outputs[1]),)
|
257 |
return (loss,) + output if loss is not None else output
|
258 |
-
return CausalLMOutputWithPast(loss=loss, logits=logits, past_key_values=outputs.past_key_values, hidden_states=outputs.hidden_states)
|
259 |
|
260 |
def param_init_fn(self, module):
|
261 |
init_fn_name = self.config.init_config['name']
|
@@ -297,4 +314,4 @@ class MPTForCausalLM(MPTPreTrainedModel):
|
|
297 |
reordered_past = []
|
298 |
for layer_past in past_key_values:
|
299 |
reordered_past += [tuple((past_state.index_select(0, beam_idx) for past_state in layer_past))]
|
300 |
-
return reordered_past
|
|
|
18 |
from .hf_prefixlm_converter import add_bidirectional_mask_if_missing, convert_hf_causal_lm_to_prefix_lm
|
19 |
from .meta_init_context import init_empty_weights
|
20 |
from .param_init_fns import MODEL_INIT_REGISTRY, generic_param_init_fn_
|
21 |
+
try:
|
22 |
+
from .flash_attn_triton import flash_attn_func
|
23 |
+
except:
|
24 |
+
pass
|
25 |
Tokenizer = Union[PreTrainedTokenizer, PreTrainedTokenizerFast]
|
26 |
|
27 |
class MPTPreTrainedModel(PreTrainedModel):
|
28 |
config_class = MPTConfig
|
29 |
base_model_prefix = 'model'
|
30 |
+
_no_split_modules = ['MPTBlock']
|
31 |
|
32 |
class MPTModel(MPTPreTrainedModel):
|
33 |
|
|
|
51 |
self.blocks = nn.ModuleList([MPTBlock(device=config.init_device, **config.to_dict()) for _ in range(config.n_layers)])
|
52 |
self.norm_f = norm_class(config.d_model, device=config.init_device)
|
53 |
if config.init_device != 'meta':
|
54 |
+
print(f'You are using config.init_device={config.init_device!r}, but you can also use config.init_device="meta" with Composer + FSDP for fast initialization.')
|
55 |
self.apply(self.param_init_fn)
|
56 |
self.is_causal = not self.prefix_lm
|
57 |
self._attn_bias_initialized = False
|
|
|
101 |
if attn_bias is None:
|
102 |
attn_bias = torch.zeros((1, 1, 1, s_k), device=device, dtype=dtype)
|
103 |
else:
|
104 |
+
_s_k = max(0, attn_bias.size(-1) - s_k)
|
105 |
+
attn_bias = attn_bias[:, :, :, _s_k:]
|
106 |
if prefix_mask is not None and attention_mask.shape != prefix_mask.shape:
|
107 |
raise ValueError(f'attention_mask shape={attention_mask.shape} ' + f'and prefix_mask shape={prefix_mask.shape} are not equal.')
|
108 |
min_val = torch.finfo(attn_bias.dtype).min
|
|
|
141 |
attention_mask = attention_mask.bool()
|
142 |
if prefix_mask is not None:
|
143 |
prefix_mask = prefix_mask.bool()
|
144 |
+
if not return_dict:
|
145 |
+
raise NotImplementedError('return_dict False is not implemented yet for MPT')
|
146 |
if output_attentions:
|
147 |
+
if self.attn_impl != 'torch':
|
148 |
+
raise NotImplementedError('output_attentions is not implemented for MPT when using attn_impl `flash` or `triton`.')
|
149 |
if attention_mask is not None and attention_mask[:, 0].sum() != attention_mask.shape[0] and self.training:
|
150 |
raise NotImplementedError('MPT does not support training with left padding.')
|
151 |
if self.prefix_lm and prefix_mask is None:
|
|
|
166 |
if len(past_key_values) != self.config.n_layers:
|
167 |
raise ValueError(f'past_key_values must provide a past_key_value for each attention ' + f'layer in the network (len(past_key_values)={len(past_key_values)!r}; self.config.n_layers={self.config.n_layers!r}).')
|
168 |
past_position = past_key_values[0][0].size(1)
|
169 |
+
if self.attn_impl == 'torch':
|
170 |
+
past_position = past_key_values[0][0].size(3)
|
171 |
if S + past_position > self.config.max_seq_len:
|
172 |
raise ValueError(f'Cannot forward input with past sequence length {past_position} and current sequence length {S + 1}, this model only supports total sequence length <= {self.config.max_seq_len}.')
|
173 |
pos = torch.arange(past_position, S + past_position, dtype=torch.long, device=input_ids.device).unsqueeze(0)
|
|
|
185 |
if use_cache and past_key_values is None:
|
186 |
past_key_values = [() for _ in range(self.config.n_layers)]
|
187 |
all_hidden_states = () if output_hidden_states else None
|
188 |
+
all_self_attns = () if output_attentions else None
|
189 |
for (b_idx, block) in enumerate(self.blocks):
|
190 |
if output_hidden_states:
|
191 |
assert all_hidden_states is not None
|
192 |
all_hidden_states = all_hidden_states + (x,)
|
193 |
past_key_value = past_key_values[b_idx] if past_key_values is not None else None
|
194 |
+
(x, attn_weights, past_key_value) = block(x, past_key_value=past_key_value, attn_bias=attn_bias, attention_mask=attention_mask, is_causal=self.is_causal)
|
195 |
if past_key_values is not None:
|
196 |
past_key_values[b_idx] = past_key_value
|
197 |
+
if output_attentions:
|
198 |
+
assert all_self_attns is not None
|
199 |
+
all_self_attns = all_self_attns + (attn_weights,)
|
200 |
x = self.norm_f(x)
|
201 |
+
if output_hidden_states:
|
202 |
+
assert all_hidden_states is not None
|
203 |
+
all_hidden_states = all_hidden_states + (x,)
|
204 |
+
if self.config.torchscript:
|
205 |
output = (x,) + (tuple(past_key_values),)
|
206 |
return output
|
207 |
+
return BaseModelOutputWithPast(last_hidden_state=x, past_key_values=past_key_values, hidden_states=all_hidden_states, attentions=all_self_attns)
|
208 |
|
209 |
def param_init_fn(self, module):
|
210 |
init_fn_name = self.config.init_config['name']
|
|
|
254 |
def forward(self, input_ids: torch.LongTensor, past_key_values: Optional[List[Tuple[torch.FloatTensor]]]=None, attention_mask: Optional[torch.ByteTensor]=None, prefix_mask: Optional[torch.ByteTensor]=None, sequence_id: Optional[torch.LongTensor]=None, labels: Optional[torch.LongTensor]=None, return_dict: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, use_cache: Optional[bool]=None):
|
255 |
return_dict = return_dict if return_dict is not None else self.config.return_dict
|
256 |
use_cache = use_cache if use_cache is not None else self.config.use_cache
|
|
|
257 |
past_key_values = list(past_key_values) if past_key_values is not None else None
|
|
|
258 |
outputs = self.transformer(input_ids=input_ids, past_key_values=past_key_values, attention_mask=attention_mask, prefix_mask=prefix_mask, sequence_id=sequence_id, return_dict=return_dict, output_attentions=output_attentions, output_hidden_states=output_hidden_states, use_cache=use_cache)
|
259 |
+
if self.config.torchscript:
|
260 |
+
logits = F.linear(outputs[0].to(self.transformer.wte.weight.device), self.transformer.wte.weight)
|
261 |
+
else:
|
262 |
+
logits = F.linear(outputs.last_hidden_state.to(self.transformer.wte.weight.device), self.transformer.wte.weight)
|
263 |
if self.logit_scale is not None:
|
264 |
if self.logit_scale == 0:
|
265 |
warnings.warn(f'Multiplying logits by self.logit_scale={self.logit_scale!r}. This will produce uniform (uninformative) outputs.')
|
|
|
269 |
labels = torch.roll(labels, shifts=-1)
|
270 |
labels[:, -1] = -100
|
271 |
loss = F.cross_entropy(logits.view(-1, logits.size(-1)), labels.to(logits.device).view(-1))
|
272 |
+
if self.config.torchscript:
|
|
|
273 |
output = (logits,) + (tuple(outputs[1]),)
|
274 |
return (loss,) + output if loss is not None else output
|
275 |
+
return CausalLMOutputWithPast(loss=loss, logits=logits, past_key_values=outputs.past_key_values, hidden_states=outputs.hidden_states, attentions=outputs.attentions)
|
276 |
|
277 |
def param_init_fn(self, module):
|
278 |
init_fn_name = self.config.init_config['name']
|
|
|
314 |
reordered_past = []
|
315 |
for layer_past in past_key_values:
|
316 |
reordered_past += [tuple((past_state.index_select(0, beam_idx) for past_state in layer_past))]
|
317 |
+
return reordered_past
|