|
"""A simple, flexible implementation of a GPT model. |
|
|
|
Inspired by https://github.com/karpathy/minGPT/blob/master/mingpt/model.py |
|
""" |
|
import math |
|
import warnings |
|
from typing import Any, Dict, List, Mapping, MutableMapping, Optional, Tuple, Union |
|
import torch |
|
import torch.nn as nn |
|
import torch.nn.functional as F |
|
from .attention import is_flash_v2_installed |
|
if is_flash_v2_installed(): |
|
try: |
|
from flash_attn.layers.rotary import RotaryEmbedding as DAILRotaryEmbedding |
|
except Exception as e: |
|
raise e |
|
from transformers import PreTrainedModel, PreTrainedTokenizerBase |
|
from transformers.modeling_outputs import BaseModelOutputWithPast, CausalLMOutputWithPast |
|
from transformers.models.llama.modeling_llama import LlamaDynamicNTKScalingRotaryEmbedding as HFDynamicNTKScalingRotaryEmbedding |
|
from transformers.models.llama.modeling_llama import LlamaLinearScalingRotaryEmbedding as HFLinearScalingRotaryEmbedding |
|
from transformers.models.llama.modeling_llama import LlamaRotaryEmbedding as HFRotaryEmbedding |
|
from .attention import ATTN_CLASS_REGISTRY, attn_bias_shape, build_attn_bias |
|
from .blocks import MPTBlock |
|
from .custom_embedding import SharedEmbedding |
|
from .fc import FC_CLASS_REGISTRY as FC_CLASS_REGISTRY |
|
from .ffn import FFN_CLASS_REGISTRY as FFN_CLASS_REGISTRY |
|
from .ffn import MPTMLP as MPTMLP |
|
from .ffn import build_ffn as build_ffn |
|
from .norm import NORM_CLASS_REGISTRY |
|
from .configuration_mpt import MPTConfig |
|
from .adapt_tokenizer import AutoTokenizerForMOD, adapt_tokenizer_for_denoising |
|
from .hf_prefixlm_converter import add_bidirectional_mask_if_missing, convert_hf_causal_lm_to_prefix_lm |
|
from .meta_init_context import init_empty_weights |
|
from .param_init_fns import generic_param_init_fn_, MODEL_INIT_REGISTRY |
|
try: |
|
from .flash_attn_triton import flash_attn_func as flash_attn_func |
|
except: |
|
pass |
|
import logging |
|
log = logging.getLogger(__name__) |
|
|
|
def gen_rotary_embedding(rope_head_dim: int, rope_impl: str, rope_theta: int, rope_dail_config: dict, rope_hf_config: dict, max_seq_len: int): |
|
if rope_impl == 'dail': |
|
return DAILRotaryEmbedding(dim=rope_head_dim, base=rope_theta, interleaved=False, scale_base=rope_dail_config['xpos_scale_base'] if rope_dail_config['type'] == 'xpos' else None, pos_idx_in_fp32=rope_dail_config['pos_idx_in_fp32'], device='cpu') |
|
elif rope_impl == 'hf': |
|
if rope_hf_config['type'] == 'no_scaling': |
|
return HFRotaryEmbedding(rope_head_dim, max_position_embeddings=max_seq_len, base=rope_theta, device='cpu') |
|
elif rope_hf_config['type'] == 'linear': |
|
return HFLinearScalingRotaryEmbedding(rope_head_dim, max_position_embeddings=max_seq_len, base=rope_theta, scaling_factor=rope_hf_config['factor'], device='cpu') |
|
elif rope_hf_config['type'] == 'dynamic': |
|
return HFDynamicNTKScalingRotaryEmbedding(rope_head_dim, max_position_embeddings=max_seq_len, base=rope_theta, scaling_factor=rope_hf_config['factor'], device='cpu') |
|
raise ValueError('rope_impl needs to be either dail or hf') |
|
|
|
class MPTPreTrainedModel(PreTrainedModel): |
|
config_class = MPTConfig |
|
base_model_prefix = 'model' |
|
_no_split_modules = ['MPTBlock'] |
|
|
|
class MPTModel(MPTPreTrainedModel): |
|
|
|
def __init__(self, config: MPTConfig): |
|
config._validate_config() |
|
super().__init__(config) |
|
self.attn_impl = config.attn_config['attn_impl'] |
|
self.prefix_lm = config.attn_config['prefix_lm'] |
|
self.attn_uses_sequence_id = config.attn_config['attn_uses_sequence_id'] |
|
self.alibi = config.attn_config['alibi'] |
|
self.alibi_bias_max = config.attn_config['alibi_bias_max'] |
|
self.learned_pos_emb = config.learned_pos_emb |
|
if config.init_device == 'mixed': |
|
if dist.get_local_rank() == 0: |
|
config.init_device = 'cpu' |
|
else: |
|
config.init_device = 'meta' |
|
if config.norm_type.lower() not in NORM_CLASS_REGISTRY.keys(): |
|
norm_options = ' | '.join(NORM_CLASS_REGISTRY.keys()) |
|
raise NotImplementedError(f'Requested norm type ({config.norm_type}) is not implemented within this repo (Options: {norm_options}).') |
|
norm_class = NORM_CLASS_REGISTRY[config.norm_type.lower()] |
|
self.embedding_fraction = config.embedding_fraction |
|
self.wte = SharedEmbedding(config.vocab_size, config.d_model, device=config.init_device) |
|
if self.learned_pos_emb: |
|
self.wpe = torch.nn.Embedding(config.max_seq_len, config.d_model, device=config.init_device) |
|
self.emb_drop = nn.Dropout(config.emb_pdrop) |
|
self.blocks = nn.ModuleList([MPTBlock(device=config.init_device, **config.to_dict()) for _ in range(config.n_layers)]) |
|
self.norm_f = norm_class(config.d_model, device=config.init_device) |
|
self.rope = config.attn_config['rope'] |
|
self.rope_impl = None |
|
if self.rope: |
|
self.rope_impl = config.attn_config['rope_impl'] |
|
self.rotary_embedding = gen_rotary_embedding(rope_head_dim=config.d_model // config.n_heads, rope_impl=self.rope_impl, rope_theta=config.attn_config['rope_theta'], rope_dail_config=config.attn_config['rope_dail_config'], rope_hf_config=config.attn_config['rope_hf_config'], max_seq_len=self.config.max_seq_len) |
|
if config.init_device != 'meta': |
|
log.info(f'We recommend using config.init_device="meta" with Composer + FSDP for faster initialization.') |
|
self.apply(self.param_init_fn) |
|
self.is_causal = not self.prefix_lm |
|
self._attn_bias_initialized = False |
|
self.attn_bias = None |
|
self.attn_bias_shape = attn_bias_shape(self.attn_impl, config.n_heads, config.max_seq_len, self.alibi, prefix_lm=self.prefix_lm, causal=self.is_causal, use_sequence_id=self.attn_uses_sequence_id) |
|
if config.no_bias: |
|
for module in self.modules(): |
|
if hasattr(module, 'bias') and isinstance(module.bias, nn.Parameter): |
|
log.info(f'Removing bias ({module.bias}) from {module}.') |
|
module.register_parameter('bias', None) |
|
if hasattr(module, 'use_bias'): |
|
log.info(f'Setting use_bias=False for {module}.') |
|
module.use_bias = False |
|
log.debug(self) |
|
log.debug(f"Using {self.config.init_config['name']} initialization.") |
|
|
|
def get_input_embeddings(self) -> Union[SharedEmbedding, nn.Embedding]: |
|
return self.wte |
|
|
|
def set_input_embeddings(self, value: Union[SharedEmbedding, nn.Embedding]) -> None: |
|
self.wte = value |
|
|
|
@torch.no_grad() |
|
def _attn_bias(self, device: torch.device, dtype: torch.dtype, attention_mask: Optional[torch.ByteTensor]=None, prefix_mask: Optional[torch.ByteTensor]=None, sequence_id: Optional[torch.LongTensor]=None) -> Tuple[Optional[torch.Tensor], Optional[torch.ByteTensor]]: |
|
if not self._attn_bias_initialized: |
|
if self.attn_bias_shape: |
|
self.attn_bias = torch.zeros(self.attn_bias_shape, device=device, dtype=dtype) |
|
self.attn_bias = build_attn_bias(self.attn_impl, self.attn_bias, self.config.n_heads, self.config.max_seq_len, causal=self.is_causal, alibi=self.alibi, alibi_bias_max=self.alibi_bias_max) |
|
self._attn_bias_initialized = True |
|
if self.attn_impl == 'flash': |
|
return (self.attn_bias, attention_mask) |
|
if self.attn_bias is not None: |
|
self.attn_bias = self.attn_bias.to(dtype=dtype, device=device) |
|
attn_bias = self.attn_bias |
|
if self.prefix_lm: |
|
assert isinstance(attn_bias, torch.Tensor) |
|
assert isinstance(prefix_mask, torch.Tensor) |
|
attn_bias = self._apply_prefix_mask(attn_bias, prefix_mask) |
|
if self.attn_uses_sequence_id and sequence_id is not None: |
|
assert isinstance(attn_bias, torch.Tensor) |
|
attn_bias = self._apply_sequence_id(attn_bias, sequence_id) |
|
if attention_mask is not None: |
|
s_k = attention_mask.shape[-1] |
|
if attn_bias is None: |
|
attn_bias = torch.zeros((1, 1, 1, s_k), device=device, dtype=dtype) |
|
else: |
|
_s_k = max(0, attn_bias.size(-1) - s_k) |
|
attn_bias = attn_bias[:, :, :, _s_k:] |
|
if prefix_mask is not None and attention_mask.shape != prefix_mask.shape: |
|
raise ValueError(f'attention_mask shape={attention_mask.shape} ' + f'and prefix_mask shape={prefix_mask.shape} are not equal.') |
|
min_val = torch.finfo(attn_bias.dtype).min |
|
attn_bias = attn_bias.masked_fill(~attention_mask.view(-1, 1, 1, s_k), min_val) |
|
return (attn_bias, None) |
|
|
|
def _apply_prefix_mask(self, attn_bias: torch.Tensor, prefix_mask: torch.Tensor) -> torch.Tensor: |
|
(s_k, s_q) = attn_bias.shape[-2:] |
|
if s_k != self.config.max_seq_len or s_q != self.config.max_seq_len: |
|
raise ValueError('attn_bias does not match the expected shape. ' + f'The last two dimensions should both be {self.config.max_length} ' + f'but are {s_k} and {s_q}.') |
|
seq_len = prefix_mask.shape[-1] |
|
if seq_len > self.config.max_seq_len: |
|
raise ValueError(f'prefix_mask sequence length cannot exceed max_seq_len={self.config.max_seq_len}') |
|
attn_bias = attn_bias[..., :seq_len, :seq_len] |
|
causal = torch.tril(torch.ones((seq_len, seq_len), dtype=torch.bool, device=prefix_mask.device)).view(1, 1, seq_len, seq_len) |
|
prefix = prefix_mask.view(-1, 1, 1, seq_len) |
|
cannot_attend = ~torch.logical_or(causal, prefix.bool()) |
|
min_val = torch.finfo(attn_bias.dtype).min |
|
attn_bias = attn_bias.masked_fill(cannot_attend, min_val) |
|
return attn_bias |
|
|
|
def _apply_sequence_id(self, attn_bias: torch.Tensor, sequence_id: torch.LongTensor) -> torch.Tensor: |
|
seq_len = sequence_id.shape[-1] |
|
if seq_len > self.config.max_seq_len: |
|
raise ValueError(f'sequence_id sequence length cannot exceed max_seq_len={self.config.max_seq_len}') |
|
attn_bias = attn_bias[..., :seq_len, :seq_len] |
|
cannot_attend = torch.logical_not(torch.eq(sequence_id.view(-1, seq_len, 1), sequence_id.view(-1, 1, seq_len))).unsqueeze(1) |
|
min_val = torch.finfo(attn_bias.dtype).min |
|
attn_bias = attn_bias.masked_fill(cannot_attend, min_val) |
|
return attn_bias |
|
|
|
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, return_dict: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, use_cache: Optional[bool]=None, inputs_embeds: Optional[torch.Tensor]=None) -> BaseModelOutputWithPast: |
|
return_dict = return_dict if return_dict is not None else self.config.return_dict |
|
use_cache = use_cache if use_cache is not None else self.config.use_cache |
|
if attention_mask is not None: |
|
attention_mask = attention_mask.bool() |
|
if prefix_mask is not None: |
|
prefix_mask = prefix_mask.bool() |
|
if not return_dict: |
|
raise NotImplementedError('return_dict False is not implemented yet for MPT') |
|
if output_attentions: |
|
if self.attn_impl != 'torch': |
|
raise NotImplementedError('output_attentions is not implemented for MPT when using attn_impl `flash` or `triton`.') |
|
if self.training and attention_mask is not None and (attention_mask[:, 0].sum() != attention_mask.shape[0]): |
|
raise NotImplementedError('MPT does not support training with left padding.') |
|
if self.prefix_lm and prefix_mask is None: |
|
raise ValueError('prefix_mask is a required argument when MPT is configured with prefix_lm=True.') |
|
if inputs_embeds is not None: |
|
raise NotImplementedError('inputs_embeds is not implemented for MPT.') |
|
if self.training: |
|
if self.attn_uses_sequence_id and sequence_id is None: |
|
raise ValueError('sequence_id is a required argument when MPT is configured with attn_uses_sequence_id=True ' + 'and the model is in train mode.') |
|
elif self.attn_uses_sequence_id is False and sequence_id is not None: |
|
warnings.warn('MPT received non-None input for `sequence_id` but is configured with attn_uses_sequence_id=False. ' + 'This input will be ignored. If you want the model to use `sequence_id`, set attn_uses_sequence_id to True.') |
|
S = input_ids.size(1) |
|
assert S <= self.config.max_seq_len, f'Cannot forward input with seq_len={S}, this model only supports seq_len<={self.config.max_seq_len}' |
|
rotary_emb_w_meta_info = None |
|
x = self.wte(input_ids) |
|
if self.learned_pos_emb or self.rope: |
|
past_position = 0 |
|
if past_key_values is not None: |
|
if len(past_key_values) != self.config.n_layers: |
|
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}).') |
|
past_position = past_key_values[0][0].size(1) |
|
if self.attn_impl == 'torch': |
|
past_position = past_key_values[0][0].size(3) |
|
if self.learned_pos_emb and S + past_position > self.config.max_seq_len: |
|
raise ValueError(f'Cannot forward input with past sequence length {past_position} and current sequence length ' + f'{S + 1}, this model only supports total sequence length <= {self.config.max_seq_len}.') |
|
if self.learned_pos_emb or (self.rope and self.rope_impl == 'hf'): |
|
pos = torch.arange(past_position, S + past_position, dtype=torch.long, device=input_ids.device).unsqueeze(0) |
|
if attention_mask is not None: |
|
pos = torch.clamp(pos - torch.cumsum((~attention_mask).to(torch.int32), dim=1)[:, past_position:], min=0) |
|
if self.learned_pos_emb: |
|
x = x + self.wpe(pos) |
|
elif self.rope and self.rope_impl == 'hf': |
|
rotary_emb_w_meta_info = {'impl': self.rope_impl, 'rotary_emb': self.rotary_embedding, 'offset_info': pos, 'seq_len': S + past_position} |
|
elif self.rope and self.rope_impl == 'dail': |
|
rotary_emb_w_meta_info = {'impl': self.rope_impl, 'rotary_emb': self.rotary_embedding, 'offset_info': past_position, 'seq_len': S + past_position} |
|
if self.embedding_fraction == 1: |
|
x = self.emb_drop(x) |
|
else: |
|
x_shrunk = x * self.embedding_fraction + x.detach() * (1 - self.embedding_fraction) |
|
assert isinstance(self.emb_drop, nn.Module) |
|
x = self.emb_drop(x_shrunk) |
|
(attn_bias, attention_mask) = self._attn_bias(device=x.device, dtype=torch.float32, attention_mask=attention_mask, prefix_mask=prefix_mask, sequence_id=sequence_id) |
|
presents = () if use_cache else None |
|
if use_cache and past_key_values is None: |
|
past_key_values = [() for _ in range(self.config.n_layers)] |
|
all_hidden_states = () if output_hidden_states else None |
|
all_self_attns = () if output_attentions else None |
|
for (b_idx, block) in enumerate(self.blocks): |
|
if output_hidden_states: |
|
assert all_hidden_states is not None |
|
all_hidden_states = all_hidden_states + (x,) |
|
past_key_value = past_key_values[b_idx] if past_key_values is not None else None |
|
(x, attn_weights, present) = block(x, past_key_value=past_key_value, attn_bias=attn_bias, rotary_emb_w_meta_info=rotary_emb_w_meta_info, attention_mask=attention_mask, is_causal=self.is_causal, output_attentions=bool(output_attentions)) |
|
if presents is not None: |
|
presents += (present,) |
|
if output_attentions: |
|
assert all_self_attns is not None |
|
all_self_attns = all_self_attns + (attn_weights,) |
|
x = self.norm_f(x) |
|
if output_hidden_states: |
|
assert all_hidden_states is not None |
|
all_hidden_states = all_hidden_states + (x,) |
|
return BaseModelOutputWithPast(last_hidden_state=x, past_key_values=presents, hidden_states=all_hidden_states, attentions=all_self_attns) |
|
|
|
def param_init_fn(self, module: nn.Module) -> None: |
|
init_fn_name = self.config.init_config['name'] |
|
MODEL_INIT_REGISTRY[init_fn_name](module=module, n_layers=self.config.n_layers, d_model=self.config.d_model, **self.config.init_config) |
|
|
|
def fsdp_wrap_fn(self, module: nn.Module) -> bool: |
|
return isinstance(module, MPTBlock) |
|
|
|
def activation_checkpointing_fn(self, module: nn.Module) -> bool: |
|
return isinstance(module, MPTBlock) |
|
|
|
class MPTForCausalLM(MPTPreTrainedModel): |
|
|
|
def __init__(self, config: MPTConfig): |
|
super().__init__(config) |
|
log.info(f'Instantiating an MPTForCausalLM model from {__file__}') |
|
self.transformer: MPTModel = MPTModel(config) |
|
self.lm_head = None |
|
if not config.tie_word_embeddings: |
|
self.lm_head = nn.Linear(config.d_model, config.vocab_size, bias=False, device=config.init_device) |
|
self.lm_head._fsdp_wrap = True |
|
for child in self.transformer.children(): |
|
if isinstance(child, torch.nn.ModuleList): |
|
continue |
|
if isinstance(child, torch.nn.Module): |
|
child._fsdp_wrap = True |
|
self.logit_scale = None |
|
if config.logit_scale is not None: |
|
logit_scale = config.logit_scale |
|
if isinstance(logit_scale, str): |
|
if logit_scale == 'inv_sqrt_d_model': |
|
logit_scale = 1 / math.sqrt(config.d_model) |
|
else: |
|
raise ValueError(f"logit_scale={logit_scale!r} is not recognized as an option; use numeric value or 'inv_sqrt_d_model'.") |
|
self.logit_scale = logit_scale |
|
|
|
def get_input_embeddings(self) -> Union[SharedEmbedding, nn.Embedding]: |
|
return self.transformer.get_input_embeddings() |
|
|
|
def set_input_embeddings(self, value: Union[SharedEmbedding, nn.Embedding]) -> None: |
|
self.transformer.set_input_embeddings(value) |
|
|
|
def get_output_embeddings(self) -> Union[SharedEmbedding, nn.Embedding, nn.Linear]: |
|
if self.lm_head is not None: |
|
return self.lm_head |
|
return self.transformer.get_input_embeddings() |
|
|
|
def set_output_embeddings(self, new_embeddings: Union[SharedEmbedding, nn.Embedding, nn.Linear]) -> None: |
|
if self.lm_head is not None: |
|
self.lm_head = new_embeddings |
|
else: |
|
if not isinstance(new_embeddings, (SharedEmbedding, nn.Embedding)): |
|
raise ValueError('new_embeddings must be an instance of SharedEmbedding ' + f'or nn.Embedding, but got {type(new_embeddings)}.') |
|
warnings.warn('Using `set_output_embeddings` to set the embedding layer of ' + 'MPTForCausalLM with tied weights. Given weights are tied, ' + 'using `set_input_embeddings` is recommended over using ' + '`set_output_embeddings`.') |
|
self.transformer.set_input_embeddings(new_embeddings) |
|
|
|
def tie_weights(self) -> None: |
|
self.lm_head = None |
|
|
|
def set_decoder(self, decoder: MPTModel) -> None: |
|
self.transformer = decoder |
|
|
|
def get_decoder(self) -> MPTModel: |
|
return self.transformer |
|
|
|
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, inputs_embeds: Optional[torch.FloatTensor]=None) -> CausalLMOutputWithPast: |
|
return_dict = return_dict if return_dict is not None else self.config.return_dict |
|
use_cache = use_cache if use_cache is not None else self.config.use_cache |
|
if inputs_embeds is not None: |
|
raise NotImplementedError('inputs_embeds has to be None (for hf/peft support).') |
|
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) |
|
if self.lm_head is not None: |
|
logits = self.lm_head(outputs.last_hidden_state) |
|
else: |
|
out = outputs.last_hidden_state |
|
out = out.to(self.transformer.wte.weight.device) |
|
logits = self.transformer.wte(out, True) |
|
if self.logit_scale is not None: |
|
if self.logit_scale == 0: |
|
warnings.warn(f'Multiplying logits by self.logit_scale={self.logit_scale!r}. This will produce uniform (uninformative) outputs.') |
|
logits *= self.logit_scale |
|
loss = None |
|
if labels is not None: |
|
_labels = torch.roll(labels, shifts=-1) |
|
_labels[:, -1] = -100 |
|
loss = F.cross_entropy(logits.view(-1, logits.size(-1)), _labels.to(logits.device).view(-1)) |
|
return CausalLMOutputWithPast(loss=loss, logits=logits, past_key_values=outputs.past_key_values, hidden_states=outputs.hidden_states, attentions=outputs.attentions) |
|
|
|
def param_init_fn(self, module: nn.Module) -> None: |
|
init_fn_name = self.config.init_config['name'] |
|
MODEL_INIT_REGISTRY[init_fn_name](module=module, n_layers=self.config.n_layers, d_model=self.config.d_model, **self.config.init_config) |
|
|
|
def fsdp_wrap_fn(self, module: nn.Module) -> bool: |
|
return isinstance(module, MPTBlock) |
|
|
|
def activation_checkpointing_fn(self, module: nn.Module) -> bool: |
|
act_ckpt_list = getattr(self.config, 'activation_checkpointing_target', None) or ['MPTBlock'] |
|
if 'MPTBlock' in act_ckpt_list or 'mptblock' in act_ckpt_list: |
|
if len(act_ckpt_list) > 1: |
|
log.info('Activation checkpointing MPTBlock only (ignoring other sub-block modules specified in activation_checkpointing_target).') |
|
return isinstance(module, MPTBlock) |
|
mod_types = () |
|
for mod_name in act_ckpt_list: |
|
if mod_name.lower() == 'mptblock': |
|
mod_types += (MPTBlock,) |
|
elif mod_name in ATTN_CLASS_REGISTRY: |
|
mod_types += (ATTN_CLASS_REGISTRY[mod_name],) |
|
elif mod_name in FFN_CLASS_REGISTRY: |
|
mod_types += (FFN_CLASS_REGISTRY[mod_name],) |
|
elif mod_name in NORM_CLASS_REGISTRY: |
|
mod_types += (NORM_CLASS_REGISTRY[mod_name],) |
|
else: |
|
msg = ', '.join(list(ATTN_CLASS_REGISTRY.keys()) + list(FFN_CLASS_REGISTRY.keys()) + list(NORM_CLASS_REGISTRY.keys()) + ['MPTBlock']) |
|
raise ValueError(f'{mod_name} (specified in activation_checkpointing_target) is not a recognized option out of available options {msg}.') |
|
return isinstance(module, mod_types) |
|
|
|
def prepare_inputs_for_generation(self, input_ids: torch.Tensor, past_key_values: Optional[List[Tuple[torch.Tensor, torch.Tensor]]]=None, inputs_embeds: Optional[torch.Tensor]=None, **kwargs: Any) -> Dict[str, Any]: |
|
if inputs_embeds is not None: |
|
raise NotImplementedError('inputs_embeds is not implemented for MPT yet') |
|
attention_mask = kwargs['attention_mask'].bool() |
|
if attention_mask[:, -1].sum() != attention_mask.shape[0]: |
|
raise NotImplementedError('MPT does not support generation with right padding.') |
|
if self.transformer.attn_uses_sequence_id and self.training: |
|
sequence_id = torch.zeros_like(input_ids[:1]) |
|
else: |
|
sequence_id = None |
|
if past_key_values is not None: |
|
input_ids = input_ids[:, -1].unsqueeze(-1) |
|
if self.transformer.prefix_lm: |
|
prefix_mask = torch.ones_like(attention_mask) |
|
if kwargs.get('use_cache') == False: |
|
raise NotImplementedError('MPT with prefix_lm=True does not support use_cache=False.') |
|
else: |
|
prefix_mask = None |
|
return {'input_ids': input_ids, 'attention_mask': attention_mask, 'prefix_mask': prefix_mask, 'sequence_id': sequence_id, 'past_key_values': past_key_values, 'use_cache': kwargs.get('use_cache', True)} |
|
|
|
@staticmethod |
|
def _reorder_cache(past_key_values: List[Tuple[torch.Tensor, torch.Tensor]], beam_idx: torch.LongTensor) -> List[Tuple[torch.Tensor, ...]]: |
|
"""Used by HuggingFace generate when using beam search with kv-caching. |
|
|
|
See https://github.com/huggingface/transformers/blob/3ec7a47664ebe40c40f4b722f6bb1cd30c3821ec/src/transformers/models/gpt2/modeling_gpt2.py#L1122-L1133 |
|
for an example in transformers. |
|
""" |
|
reordered_past = [] |
|
for layer_past in past_key_values: |
|
reordered_past += [tuple((past_state.index_select(0, beam_idx) for past_state in layer_past))] |
|
return reordered_past |
