初始化项目，由ModelHub XC社区提供模型
Model: TongjiFinLab/CFGPT1-pt-7B Source: Original Platform
This commit is contained in:
ModelHub XC
35
.gitattributes
vendored
Normal file
35
.gitattributes
vendored
Normal file
@@ -0,0 +1,35 @@
|
||||
*.7z filter=lfs diff=lfs merge=lfs -text
|
||||
*.arrow filter=lfs diff=lfs merge=lfs -text
|
||||
*.bin filter=lfs diff=lfs merge=lfs -text
|
||||
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
||||
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
||||
*.ftz filter=lfs diff=lfs merge=lfs -text
|
||||
*.gz filter=lfs diff=lfs merge=lfs -text
|
||||
*.h5 filter=lfs diff=lfs merge=lfs -text
|
||||
*.joblib filter=lfs diff=lfs merge=lfs -text
|
||||
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
||||
*.model filter=lfs diff=lfs merge=lfs -text
|
||||
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
||||
*.npy filter=lfs diff=lfs merge=lfs -text
|
||||
*.npz filter=lfs diff=lfs merge=lfs -text
|
||||
*.onnx filter=lfs diff=lfs merge=lfs -text
|
||||
*.ot filter=lfs diff=lfs merge=lfs -text
|
||||
*.parquet filter=lfs diff=lfs merge=lfs -text
|
||||
*.pb filter=lfs diff=lfs merge=lfs -text
|
||||
*.pickle filter=lfs diff=lfs merge=lfs -text
|
||||
*.pkl filter=lfs diff=lfs merge=lfs -text
|
||||
*.pt filter=lfs diff=lfs merge=lfs -text
|
||||
*.pth filter=lfs diff=lfs merge=lfs -text
|
||||
*.rar filter=lfs diff=lfs merge=lfs -text
|
||||
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
||||
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar filter=lfs diff=lfs merge=lfs -text
|
||||
*.tflite filter=lfs diff=lfs merge=lfs -text
|
||||
*.tgz filter=lfs diff=lfs merge=lfs -text
|
||||
*.wasm filter=lfs diff=lfs merge=lfs -text
|
||||
*.xz filter=lfs diff=lfs merge=lfs -text
|
||||
*.zip filter=lfs diff=lfs merge=lfs -text
|
||||
*.zst filter=lfs diff=lfs merge=lfs -text
|
||||
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
||||
40
README.md
Normal file
40
README.md
Normal file
@@ -0,0 +1,40 @@
|
||||
---
|
||||
license: apache-2.0
|
||||
language:
|
||||
- zh
|
||||
pipeline_tag: text-generation
|
||||
---
|
||||
---
|
||||
<div style="text-align:center">
|
||||
<!-- <img src="https://big-cheng.com/k2/k2.png" alt="k2-logo" width="200"/> -->
|
||||
<h2>📈 CFGPT: Chinese Financial Assistant with Large Language Model (CFGPT1-pt-7b)</h2>
|
||||
</div>
|
||||
|
||||
## Introduction
|
||||
|
||||
We introduce **CFGPT**, an open-source language model trained by firstly further pretraining general LLMs on collected and cleaned Chinese finance text data (CFData-pt), including financial domain-specific data (announcement, finance articles, finance exams, finance news, finance research papers) and general data (Wikipedia), and secondly fine-tuning with knowledge-intensive instruction tuning data (CFData-sft).
|
||||
As for preliminary evaluation, we use CFBenchmark-Basic.
|
||||
CFGPT outperforms the baselines on objective and subjective tasks compared to several baseline models with similar parameters.
|
||||
|
||||
In this repository, we will share the further pretrained model.
|
||||
|
||||
- [Pretrained Model](https://huggingface.co/TongjiFinLab/CFGPT1-pt-7B): Full model weights after further pretraining with the chinese finance text corpus to comply with the InternLM model license.
|
||||
|
||||
## How to Use
|
||||
|
||||
The CFGPT1-pt-7b is a pre-trained model, which has not undergone supervised fine-tuning with a instruction data. Therefore, it is not advisable to use this model for financial tasks.
|
||||
Please refer to [CFGPT]() Github repo for further usage.
|
||||
|
||||
## 简介
|
||||
|
||||
**CFGPT**是一个开源的语言模型，首先通过在收集和清理的中国金融文本数据（CFData-pt）上进行继续预训练，包括金融领域特定数据（公告、金融文章、金融考试、金融新闻、金融研究论文）和通用数据（维基百科），然后使用知识密集的指导调整数据（CFData-sft）进行微调。
|
||||
我们使用CFBenchmark-Basic进行初步评估。与几个具有相似参数的基线模型相比，CFGPT在识别，分类和生成任务上表现优越。
|
||||
|
||||
在这个仓库中，我们将分享以下继续预训练的模型。
|
||||
|
||||
- [Pretrained Model](https://huggingface.co/TongjiFinLab/CFGPT1-pt-7B): 在中国金融文本语料库上进行进一步预训练且符合InternLM模型许可的完整模型权重。
|
||||
|
||||
## 如何使用
|
||||
|
||||
这个模型是一个预训练的模型，还没有经历过指令数据库的有监督微调，因此不建议使用该模型执行相关金融任务。
|
||||
具体使用，请参考[CFGPT]()的Github仓库。
|
||||
5
added_tokens.json
Normal file
5
added_tokens.json
Normal file
@@ -0,0 +1,5 @@
|
||||
{
|
||||
"</s>": 2,
|
||||
"<s>": 1,
|
||||
"<unk>": 0
|
||||
}
|
||||
29
config.json
Normal file
29
config.json
Normal file
@@ -0,0 +1,29 @@
|
||||
{
|
||||
"_name_or_path": "/mnt/workspace/hf_model_release/CFGPT-pt-7b",
|
||||
"architectures": [
|
||||
"InternLMForCausalLM"
|
||||
],
|
||||
"auto_map": {
|
||||
"AutoConfig": "configuration_internlm.InternLMConfig",
|
||||
"AutoModel": "modeling_internlm.InternLMForCausalLM",
|
||||
"AutoModelForCausalLM": "modeling_internlm.InternLMForCausalLM"
|
||||
},
|
||||
"bias": true,
|
||||
"bos_token_id": 1,
|
||||
"eos_token_id": 2,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 4096,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 11008,
|
||||
"max_position_embeddings": 2048,
|
||||
"model_type": "internlm",
|
||||
"num_attention_heads": 32,
|
||||
"num_hidden_layers": 32,
|
||||
"pad_token_id": 0,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"tie_word_embeddings": false,
|
||||
"torch_dtype": "bfloat16",
|
||||
"transformers_version": "4.34.0",
|
||||
"use_cache": false,
|
||||
"vocab_size": 103168
|
||||
}
|
||||
120
configuration_internlm.py
Normal file
120
configuration_internlm.py
Normal file
@@ -0,0 +1,120 @@
|
||||
# coding=utf-8
|
||||
# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
|
||||
#
|
||||
# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
|
||||
# and OPT implementations in this library. It has been modified from its
|
||||
# original forms to accommodate minor architectural differences compared
|
||||
# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
""" InternLM model configuration"""
|
||||
|
||||
from transformers.utils import logging
|
||||
from transformers.configuration_utils import PretrainedConfig
|
||||
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
INTERNLM_PRETRAINED_CONFIG_ARCHIVE_MAP = {}
|
||||
|
||||
|
||||
class InternLMConfig(PretrainedConfig):
|
||||
r"""
|
||||
This is the configuration class to store the configuration of a [`InternLMModel`]. It is used to instantiate an InternLM
|
||||
model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
|
||||
defaults will yield a similar configuration to that of the InternLM-7B.
|
||||
|
||||
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
|
||||
documentation from [`PretrainedConfig`] for more information.
|
||||
|
||||
|
||||
Args:
|
||||
vocab_size (`int`, *optional*, defaults to 32000):
|
||||
Vocabulary size of the InternLM model. Defines the number of different tokens that can be represented by the
|
||||
`inputs_ids` passed when calling [`InternLMModel`]
|
||||
hidden_size (`int`, *optional*, defaults to 4096):
|
||||
Dimension of the hidden representations.
|
||||
intermediate_size (`int`, *optional*, defaults to 11008):
|
||||
Dimension of the MLP representations.
|
||||
num_hidden_layers (`int`, *optional*, defaults to 32):
|
||||
Number of hidden layers in the Transformer encoder.
|
||||
num_attention_heads (`int`, *optional*, defaults to 32):
|
||||
Number of attention heads for each attention layer in the Transformer encoder.
|
||||
hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
|
||||
The non-linear activation function (function or string) in the decoder.
|
||||
max_position_embeddings (`int`, *optional*, defaults to 2048):
|
||||
The maximum sequence length that this model might ever be used with. Typically set this to something large
|
||||
just in case (e.g., 512 or 1024 or 2048).
|
||||
initializer_range (`float`, *optional*, defaults to 0.02):
|
||||
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
|
||||
rms_norm_eps (`float`, *optional*, defaults to 1e-12):
|
||||
The epsilon used by the rms normalization layers.
|
||||
use_cache (`bool`, *optional*, defaults to `True`):
|
||||
Whether or not the model should return the last key/values attentions (not used by all models). Only
|
||||
relevant if `config.is_decoder=True`.
|
||||
tie_word_embeddings(`bool`, *optional*, defaults to `False`):
|
||||
Whether to tie weight embeddings
|
||||
Example:
|
||||
|
||||
```python
|
||||
>>> from transformers import InternLMModel, InternLMConfig
|
||||
|
||||
>>> # Initializing a InternLM internlm-7b style configuration
|
||||
>>> configuration = InternLMConfig()
|
||||
|
||||
>>> # Initializing a model from the internlm-7b style configuration
|
||||
>>> model = InternLMModel(configuration)
|
||||
|
||||
>>> # Accessing the model configuration
|
||||
>>> configuration = model.config
|
||||
```"""
|
||||
model_type = "internlm"
|
||||
_auto_class = "AutoConfig"
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
vocab_size=103168,
|
||||
hidden_size=4096,
|
||||
intermediate_size=11008,
|
||||
num_hidden_layers=32,
|
||||
num_attention_heads=32,
|
||||
hidden_act="silu",
|
||||
max_position_embeddings=2048,
|
||||
initializer_range=0.02,
|
||||
rms_norm_eps=1e-6,
|
||||
use_cache=True,
|
||||
pad_token_id=0,
|
||||
bos_token_id=1,
|
||||
eos_token_id=2,
|
||||
tie_word_embeddings=False,
|
||||
bias=True,
|
||||
**kwargs,
|
||||
):
|
||||
self.vocab_size = vocab_size
|
||||
self.max_position_embeddings = max_position_embeddings
|
||||
self.hidden_size = hidden_size
|
||||
self.intermediate_size = intermediate_size
|
||||
self.num_hidden_layers = num_hidden_layers
|
||||
self.num_attention_heads = num_attention_heads
|
||||
self.hidden_act = hidden_act
|
||||
self.initializer_range = initializer_range
|
||||
self.rms_norm_eps = rms_norm_eps
|
||||
self.use_cache = use_cache
|
||||
self.bias = bias
|
||||
super().__init__(
|
||||
pad_token_id=pad_token_id,
|
||||
bos_token_id=bos_token_id,
|
||||
eos_token_id=eos_token_id,
|
||||
tie_word_embeddings=tie_word_embeddings,
|
||||
**kwargs,
|
||||
)
|
||||
7
generation_config.json
Normal file
7
generation_config.json
Normal file
@@ -0,0 +1,7 @@
|
||||
{
|
||||
"_from_model_config": true,
|
||||
"bos_token_id": 1,
|
||||
"eos_token_id": 2,
|
||||
"pad_token_id": 0,
|
||||
"transformers_version": "4.34.0"
|
||||
}
|
||||
996
modeling_internlm.py
Normal file
996
modeling_internlm.py
Normal file
@@ -0,0 +1,996 @@
|
||||
# coding=utf-8
|
||||
# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
|
||||
#
|
||||
# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
|
||||
# and OPT implementations in this library. It has been modified from its
|
||||
# original forms to accommodate minor architectural differences compared
|
||||
# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
""" PyTorch InternLM model."""
|
||||
import math
|
||||
from typing import List, Optional, Tuple, Union
|
||||
import threading, queue
|
||||
|
||||
import torch
|
||||
import torch.utils.checkpoint
|
||||
from torch import nn
|
||||
from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
|
||||
|
||||
from transformers.activations import ACT2FN
|
||||
from transformers.modeling_outputs import BaseModelOutputWithPast, CausalLMOutputWithPast, SequenceClassifierOutputWithPast
|
||||
from transformers.modeling_utils import PreTrainedModel
|
||||
from transformers.generation.streamers import BaseStreamer
|
||||
from transformers.utils import add_start_docstrings, add_start_docstrings_to_model_forward, logging, replace_return_docstrings
|
||||
from .configuration_internlm import InternLMConfig
|
||||
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
_CONFIG_FOR_DOC = "InternLMConfig"
|
||||
|
||||
# Copied from transformers.models.bart.modeling_bart._make_causal_mask
|
||||
def _make_causal_mask(
|
||||
input_ids_shape: torch.Size, dtype: torch.dtype, device: torch.device, past_key_values_length: int = 0
|
||||
):
|
||||
"""
|
||||
Make causal mask used for bi-directional self-attention.
|
||||
"""
|
||||
bsz, tgt_len = input_ids_shape
|
||||
mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)
|
||||
mask_cond = torch.arange(mask.size(-1), device=device)
|
||||
mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
|
||||
mask = mask.to(dtype)
|
||||
|
||||
if past_key_values_length > 0:
|
||||
mask = torch.cat([torch.zeros(tgt_len, past_key_values_length, dtype=dtype, device=device), mask], dim=-1)
|
||||
return mask[None, None, :, :].expand(bsz, 1, tgt_len, tgt_len + past_key_values_length)
|
||||
|
||||
|
||||
# Copied from transformers.models.bart.modeling_bart._expand_mask
|
||||
def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Optional[int] = None):
|
||||
"""
|
||||
Expands attention_mask from `[bsz, seq_len]` to `[bsz, 1, tgt_seq_len, src_seq_len]`.
|
||||
"""
|
||||
bsz, src_len = mask.size()
|
||||
tgt_len = tgt_len if tgt_len is not None else src_len
|
||||
|
||||
expanded_mask = mask[:, None, None, :].expand(bsz, 1, tgt_len, src_len).to(dtype)
|
||||
|
||||
inverted_mask = 1.0 - expanded_mask
|
||||
|
||||
return inverted_mask.masked_fill(inverted_mask.to(torch.bool), torch.finfo(dtype).min)
|
||||
|
||||
|
||||
class InternLMRMSNorm(nn.Module):
|
||||
def __init__(self, hidden_size, eps=1e-6):
|
||||
"""
|
||||
InternLMRMSNorm is equivalent to T5LayerNorm
|
||||
"""
|
||||
super().__init__()
|
||||
self.weight = nn.Parameter(torch.ones(hidden_size))
|
||||
self.variance_epsilon = eps
|
||||
|
||||
def forward(self, hidden_states):
|
||||
variance = hidden_states.to(torch.float32).pow(2).mean(-1, keepdim=True)
|
||||
hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon)
|
||||
|
||||
# convert into half-precision if necessary
|
||||
if self.weight.dtype in [torch.float16, torch.bfloat16]:
|
||||
hidden_states = hidden_states.to(self.weight.dtype)
|
||||
|
||||
return self.weight * hidden_states
|
||||
|
||||
|
||||
class InternLMRotaryEmbedding(torch.nn.Module):
|
||||
def __init__(self, dim, max_position_embeddings=2048, base=10000, device=None):
|
||||
super().__init__()
|
||||
inv_freq = 1.0 / (base ** (torch.arange(0, dim, 2).float().to(device) / dim))
|
||||
self.register_buffer("inv_freq", inv_freq, persistent=False)
|
||||
|
||||
# Build here to make `torch.jit.trace` work.
|
||||
self.max_seq_len_cached = max_position_embeddings
|
||||
t = torch.arange(self.max_seq_len_cached, device=self.inv_freq.device, dtype=self.inv_freq.dtype)
|
||||
freqs = torch.einsum("i,j->ij", t, self.inv_freq)
|
||||
# Different from paper, but it uses a different permutation in order to obtain the same calculation
|
||||
emb = torch.cat((freqs, freqs), dim=-1)
|
||||
self.register_buffer("cos_cached", emb.cos()[None, None, :, :], persistent=False)
|
||||
self.register_buffer("sin_cached", emb.sin()[None, None, :, :], persistent=False)
|
||||
|
||||
def forward(self, x, seq_len=None):
|
||||
# x: [bs, num_attention_heads, seq_len, head_size]
|
||||
# This `if` block is unlikely to be run after we build sin/cos in `__init__`. Keep the logic here just in case.
|
||||
if seq_len > self.max_seq_len_cached:
|
||||
self.max_seq_len_cached = seq_len
|
||||
t = torch.arange(self.max_seq_len_cached, device=x.device, dtype=self.inv_freq.dtype)
|
||||
freqs = torch.einsum("i,j->ij", t, self.inv_freq)
|
||||
# Different from paper, but it uses a different permutation in order to obtain the same calculation
|
||||
emb = torch.cat((freqs, freqs), dim=-1).to(x.device)
|
||||
self.register_buffer("cos_cached", emb.cos()[None, None, :, :], persistent=False)
|
||||
self.register_buffer("sin_cached", emb.sin()[None, None, :, :], persistent=False)
|
||||
return (
|
||||
self.cos_cached[:, :, :seq_len, ...].to(dtype=x.dtype),
|
||||
self.sin_cached[:, :, :seq_len, ...].to(dtype=x.dtype),
|
||||
)
|
||||
|
||||
|
||||
def rotate_half(x):
|
||||
"""Rotates half the hidden dims of the input."""
|
||||
x1 = x[..., : x.shape[-1] // 2]
|
||||
x2 = x[..., x.shape[-1] // 2 :]
|
||||
return torch.cat((-x2, x1), dim=-1)
|
||||
|
||||
|
||||
def apply_rotary_pos_emb(q, k, cos, sin, position_ids):
|
||||
# The first two dimensions of cos and sin are always 1, so we can `squeeze` them.
|
||||
cos = cos.squeeze(1).squeeze(0) # [seq_len, dim]
|
||||
sin = sin.squeeze(1).squeeze(0) # [seq_len, dim]
|
||||
cos = cos[position_ids].unsqueeze(1) # [bs, 1, seq_len, dim]
|
||||
sin = sin[position_ids].unsqueeze(1) # [bs, 1, seq_len, dim]
|
||||
q_embed = (q * cos) + (rotate_half(q) * sin)
|
||||
k_embed = (k * cos) + (rotate_half(k) * sin)
|
||||
return q_embed, k_embed
|
||||
|
||||
|
||||
class InternLMMLP(nn.Module):
|
||||
def __init__(
|
||||
self,
|
||||
hidden_size: int,
|
||||
intermediate_size: int,
|
||||
hidden_act: str,
|
||||
):
|
||||
super().__init__()
|
||||
self.gate_proj = nn.Linear(hidden_size, intermediate_size, bias=False)
|
||||
self.down_proj = nn.Linear(intermediate_size, hidden_size, bias=False)
|
||||
self.up_proj = nn.Linear(hidden_size, intermediate_size, bias=False)
|
||||
self.act_fn = ACT2FN[hidden_act]
|
||||
|
||||
def forward(self, x):
|
||||
return self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
|
||||
|
||||
|
||||
class InternLMAttention(nn.Module):
|
||||
"""Multi-headed attention from 'Attention Is All You Need' paper"""
|
||||
|
||||
def __init__(self, config: InternLMConfig):
|
||||
super().__init__()
|
||||
self.config = config
|
||||
self.hidden_size = config.hidden_size
|
||||
self.num_heads = config.num_attention_heads
|
||||
self.head_dim = self.hidden_size // self.num_heads
|
||||
self.max_position_embeddings = config.max_position_embeddings
|
||||
|
||||
if (self.head_dim * self.num_heads) != self.hidden_size:
|
||||
raise ValueError(
|
||||
f"hidden_size must be divisible by num_heads (got `hidden_size`: {self.hidden_size}"
|
||||
f" and `num_heads`: {self.num_heads})."
|
||||
)
|
||||
self.q_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=config.bias)
|
||||
self.k_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=config.bias)
|
||||
self.v_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=config.bias)
|
||||
self.o_proj = nn.Linear(self.num_heads * self.head_dim, self.hidden_size, bias=config.bias)
|
||||
self.rotary_emb = InternLMRotaryEmbedding(self.head_dim, max_position_embeddings=self.max_position_embeddings)
|
||||
|
||||
def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
|
||||
return tensor.view(bsz, seq_len, self.num_heads, self.head_dim).transpose(1, 2).contiguous()
|
||||
|
||||
def forward(
|
||||
self,
|
||||
hidden_states: torch.Tensor,
|
||||
attention_mask: Optional[torch.Tensor] = None,
|
||||
position_ids: Optional[torch.LongTensor] = None,
|
||||
past_key_value: Optional[Tuple[torch.Tensor]] = None,
|
||||
output_attentions: bool = False,
|
||||
use_cache: bool = False,
|
||||
) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
|
||||
bsz, q_len, _ = hidden_states.size()
|
||||
|
||||
query_states = self.q_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
|
||||
key_states = self.k_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
|
||||
value_states = self.v_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
|
||||
|
||||
kv_seq_len = key_states.shape[-2]
|
||||
if past_key_value is not None:
|
||||
kv_seq_len += past_key_value[0].shape[-2]
|
||||
cos, sin = self.rotary_emb(value_states, seq_len=kv_seq_len)
|
||||
query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin, position_ids)
|
||||
# [bsz, nh, t, hd]
|
||||
|
||||
if past_key_value is not None:
|
||||
# reuse k, v, self_attention
|
||||
key_states = torch.cat([past_key_value[0], key_states], dim=2)
|
||||
value_states = torch.cat([past_key_value[1], value_states], dim=2)
|
||||
|
||||
past_key_value = (key_states, value_states) if use_cache else None
|
||||
|
||||
attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) / math.sqrt(self.head_dim)
|
||||
|
||||
if attn_weights.size() != (bsz, self.num_heads, q_len, kv_seq_len):
|
||||
raise ValueError(
|
||||
f"Attention weights should be of size {(bsz, self.num_heads, q_len, kv_seq_len)}, but is"
|
||||
f" {attn_weights.size()}"
|
||||
)
|
||||
|
||||
if attention_mask is not None:
|
||||
if attention_mask.size() != (bsz, 1, q_len, kv_seq_len):
|
||||
raise ValueError(
|
||||
f"Attention mask should be of size {(bsz, 1, q_len, kv_seq_len)}, but is {attention_mask.size()}"
|
||||
)
|
||||
attn_weights = attn_weights + attention_mask
|
||||
attn_weights = torch.max(attn_weights, torch.tensor(torch.finfo(attn_weights.dtype).min))
|
||||
|
||||
# upcast attention to fp32
|
||||
attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query_states.dtype)
|
||||
attn_output = torch.matmul(attn_weights, value_states)
|
||||
|
||||
if attn_output.size() != (bsz, self.num_heads, q_len, self.head_dim):
|
||||
raise ValueError(
|
||||
f"`attn_output` should be of size {(bsz, self.num_heads, q_len, self.head_dim)}, but is"
|
||||
f" {attn_output.size()}"
|
||||
)
|
||||
|
||||
attn_output = attn_output.transpose(1, 2)
|
||||
attn_output = attn_output.reshape(bsz, q_len, self.hidden_size)
|
||||
|
||||
attn_output = self.o_proj(attn_output)
|
||||
|
||||
if not output_attentions:
|
||||
attn_weights = None
|
||||
|
||||
return attn_output, attn_weights, past_key_value
|
||||
|
||||
|
||||
class InternLMDecoderLayer(nn.Module):
|
||||
def __init__(self, config: InternLMConfig):
|
||||
super().__init__()
|
||||
self.hidden_size = config.hidden_size
|
||||
self.self_attn = InternLMAttention(config=config)
|
||||
self.mlp = InternLMMLP(
|
||||
hidden_size=self.hidden_size,
|
||||
intermediate_size=config.intermediate_size,
|
||||
hidden_act=config.hidden_act,
|
||||
)
|
||||
self.input_layernorm = InternLMRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
|
||||
self.post_attention_layernorm = InternLMRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
|
||||
|
||||
def forward(
|
||||
self,
|
||||
hidden_states: torch.Tensor,
|
||||
attention_mask: Optional[torch.Tensor] = None,
|
||||
position_ids: Optional[torch.LongTensor] = None,
|
||||
past_key_value: Optional[Tuple[torch.Tensor]] = None,
|
||||
output_attentions: Optional[bool] = False,
|
||||
use_cache: Optional[bool] = False,
|
||||
) -> Tuple[torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]]:
|
||||
"""
|
||||
Args:
|
||||
hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
|
||||
attention_mask (`torch.FloatTensor`, *optional*): attention mask of size
|
||||
`(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
|
||||
output_attentions (`bool`, *optional*):
|
||||
Whether or not to return the attentions tensors of all attention layers. See `attentions` under
|
||||
returned tensors for more detail.
|
||||
use_cache (`bool`, *optional*):
|
||||
If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
|
||||
(see `past_key_values`).
|
||||
past_key_value (`Tuple(torch.FloatTensor)`, *optional*): cached past key and value projection states
|
||||
"""
|
||||
|
||||
residual = hidden_states
|
||||
|
||||
hidden_states = self.input_layernorm(hidden_states)
|
||||
|
||||
# Self Attention
|
||||
hidden_states, self_attn_weights, present_key_value = self.self_attn(
|
||||
hidden_states=hidden_states,
|
||||
attention_mask=attention_mask,
|
||||
position_ids=position_ids,
|
||||
past_key_value=past_key_value,
|
||||
output_attentions=output_attentions,
|
||||
use_cache=use_cache,
|
||||
)
|
||||
hidden_states = residual + hidden_states
|
||||
|
||||
# Fully Connected
|
||||
residual = hidden_states
|
||||
hidden_states = self.post_attention_layernorm(hidden_states)
|
||||
hidden_states = self.mlp(hidden_states)
|
||||
hidden_states = residual + hidden_states
|
||||
|
||||
outputs = (hidden_states,)
|
||||
|
||||
if output_attentions:
|
||||
outputs += (self_attn_weights,)
|
||||
|
||||
if use_cache:
|
||||
outputs += (present_key_value,)
|
||||
|
||||
return outputs
|
||||
|
||||
|
||||
INTERNLM_START_DOCSTRING = r"""
|
||||
This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
|
||||
library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
|
||||
etc.)
|
||||
|
||||
This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
|
||||
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
|
||||
and behavior.
|
||||
|
||||
Parameters:
|
||||
config ([`InternLMConfig`]):
|
||||
Model configuration class with all the parameters of the model. Initializing with a config file does not
|
||||
load the weights associated with the model, only the configuration. Check out the
|
||||
[`~PreTrainedModel.from_pretrained`] method to load the model weights.
|
||||
"""
|
||||
|
||||
|
||||
@add_start_docstrings(
|
||||
"The bare InternLM Model outputting raw hidden-states without any specific head on top.",
|
||||
INTERNLM_START_DOCSTRING,
|
||||
)
|
||||
class InternLMPreTrainedModel(PreTrainedModel):
|
||||
config_class = InternLMConfig
|
||||
base_model_prefix = "model"
|
||||
supports_gradient_checkpointing = True
|
||||
_no_split_modules = ["InternLMDecoderLayer"]
|
||||
_keys_to_ignore_on_load_unexpected = [r"decoder\.version"]
|
||||
|
||||
def _init_weights(self, module):
|
||||
std = self.config.initializer_range
|
||||
if isinstance(module, nn.Linear):
|
||||
module.weight.data.normal_(mean=0.0, std=std)
|
||||
if module.bias is not None:
|
||||
module.bias.data.zero_()
|
||||
elif isinstance(module, nn.Embedding):
|
||||
module.weight.data.normal_(mean=0.0, std=std)
|
||||
if module.padding_idx is not None:
|
||||
module.weight.data[module.padding_idx].zero_()
|
||||
|
||||
def _set_gradient_checkpointing(self, module, value=False):
|
||||
if isinstance(module, InternLMModel):
|
||||
module.gradient_checkpointing = value
|
||||
|
||||
|
||||
INTERNLM_INPUTS_DOCSTRING = r"""
|
||||
Args:
|
||||
input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
|
||||
Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
|
||||
it.
|
||||
|
||||
Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
|
||||
[`PreTrainedTokenizer.__call__`] for details.
|
||||
|
||||
[What are input IDs?](../glossary#input-ids)
|
||||
attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
|
||||
Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
|
||||
|
||||
- 1 for tokens that are **not masked**,
|
||||
- 0 for tokens that are **masked**.
|
||||
|
||||
[What are attention masks?](../glossary#attention-mask)
|
||||
|
||||
Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
|
||||
[`PreTrainedTokenizer.__call__`] for details.
|
||||
|
||||
If `past_key_values` is used, optionally only the last `decoder_input_ids` have to be input (see
|
||||
`past_key_values`).
|
||||
|
||||
If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`]
|
||||
and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
|
||||
information on the default strategy.
|
||||
|
||||
- 1 indicates the head is **not masked**,
|
||||
- 0 indicates the head is **masked**.
|
||||
position_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
|
||||
Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
|
||||
config.n_positions - 1]`.
|
||||
|
||||
[What are position IDs?](../glossary#position-ids)
|
||||
past_key_values (`tuple(tuple(torch.FloatTensor))`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`):
|
||||
Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of shape
|
||||
`(batch_size, num_heads, sequence_length, embed_size_per_head)`) and 2 additional tensors of shape
|
||||
`(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)`.
|
||||
|
||||
Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
|
||||
blocks) that can be used (see `past_key_values` input) to speed up sequential decoding.
|
||||
|
||||
If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
|
||||
don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
|
||||
`decoder_input_ids` of shape `(batch_size, sequence_length)`.
|
||||
inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
|
||||
Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
|
||||
is useful if you want more control over how to convert `input_ids` indices into associated vectors than the
|
||||
model's internal embedding lookup matrix.
|
||||
use_cache (`bool`, *optional*):
|
||||
If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
|
||||
`past_key_values`).
|
||||
output_attentions (`bool`, *optional*):
|
||||
Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned
|
||||
tensors for more detail.
|
||||
output_hidden_states (`bool`, *optional*):
|
||||
Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
|
||||
more detail.
|
||||
return_dict (`bool`, *optional*):
|
||||
Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
|
||||
"""
|
||||
|
||||
|
||||
@add_start_docstrings(
|
||||
"The bare InternLM Model outputting raw hidden-states without any specific head on top.",
|
||||
INTERNLM_START_DOCSTRING,
|
||||
)
|
||||
class InternLMModel(InternLMPreTrainedModel):
|
||||
"""
|
||||
Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`InternLMDecoderLayer`]
|
||||
|
||||
Args:
|
||||
config: InternLMConfig
|
||||
"""
|
||||
_auto_class = "AutoModel"
|
||||
|
||||
def __init__(self, config: InternLMConfig):
|
||||
super().__init__(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, self.padding_idx)
|
||||
self.layers = nn.ModuleList([InternLMDecoderLayer(config) for _ in range(config.num_hidden_layers)])
|
||||
self.norm = InternLMRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
|
||||
|
||||
self.gradient_checkpointing = False
|
||||
# Initialize weights and apply final processing
|
||||
self.post_init()
|
||||
|
||||
def get_input_embeddings(self):
|
||||
return self.embed_tokens
|
||||
|
||||
def set_input_embeddings(self, value):
|
||||
self.embed_tokens = value
|
||||
|
||||
# Copied from transformers.models.bart.modeling_bart.BartDecoder._prepare_decoder_attention_mask
|
||||
def _prepare_decoder_attention_mask(self, attention_mask, input_shape, inputs_embeds, past_key_values_length):
|
||||
# create causal mask
|
||||
# [bsz, seq_len] -> [bsz, 1, tgt_seq_len, src_seq_len]
|
||||
combined_attention_mask = None
|
||||
if input_shape[-1] > 1:
|
||||
combined_attention_mask = _make_causal_mask(
|
||||
input_shape,
|
||||
inputs_embeds.dtype,
|
||||
device=inputs_embeds.device,
|
||||
past_key_values_length=past_key_values_length,
|
||||
)
|
||||
|
||||
if attention_mask is not None:
|
||||
# [bsz, seq_len] -> [bsz, 1, tgt_seq_len, src_seq_len]
|
||||
expanded_attn_mask = _expand_mask(attention_mask, inputs_embeds.dtype, tgt_len=input_shape[-1]).to(
|
||||
inputs_embeds.device
|
||||
)
|
||||
combined_attention_mask = (
|
||||
expanded_attn_mask if combined_attention_mask is None else expanded_attn_mask + combined_attention_mask
|
||||
)
|
||||
|
||||
return combined_attention_mask
|
||||
|
||||
@add_start_docstrings_to_model_forward(INTERNLM_INPUTS_DOCSTRING)
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.LongTensor = None,
|
||||
attention_mask: Optional[torch.Tensor] = None,
|
||||
position_ids: Optional[torch.LongTensor] = None,
|
||||
past_key_values: Optional[List[torch.FloatTensor]] = None,
|
||||
inputs_embeds: Optional[torch.FloatTensor] = None,
|
||||
use_cache: Optional[bool] = None,
|
||||
output_attentions: Optional[bool] = None,
|
||||
output_hidden_states: Optional[bool] = None,
|
||||
return_dict: Optional[bool] = None,
|
||||
) -> Union[Tuple, BaseModelOutputWithPast]:
|
||||
output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
|
||||
output_hidden_states = (
|
||||
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
|
||||
)
|
||||
use_cache = use_cache if use_cache is not None else self.config.use_cache
|
||||
|
||||
return_dict = return_dict if return_dict is not None else self.config.use_return_dict
|
||||
|
||||
# retrieve input_ids and inputs_embeds
|
||||
if input_ids is not None and inputs_embeds is not None:
|
||||
raise ValueError("You cannot specify both decoder_input_ids and decoder_inputs_embeds at the same time")
|
||||
elif input_ids is not None:
|
||||
batch_size, seq_length = input_ids.shape
|
||||
elif inputs_embeds is not None:
|
||||
batch_size, seq_length, _ = inputs_embeds.shape
|
||||
else:
|
||||
raise ValueError("You have to specify either decoder_input_ids or decoder_inputs_embeds")
|
||||
|
||||
seq_length_with_past = seq_length
|
||||
past_key_values_length = 0
|
||||
|
||||
if past_key_values is not None:
|
||||
past_key_values_length = past_key_values[0][0].shape[2]
|
||||
seq_length_with_past = seq_length_with_past + past_key_values_length
|
||||
|
||||
if position_ids is None:
|
||||
device = input_ids.device if input_ids is not None else inputs_embeds.device
|
||||
position_ids = torch.arange(
|
||||
past_key_values_length, seq_length + past_key_values_length, dtype=torch.long, device=device
|
||||
)
|
||||
position_ids = position_ids.unsqueeze(0).view(-1, seq_length)
|
||||
else:
|
||||
position_ids = position_ids.view(-1, seq_length).long()
|
||||
|
||||
if inputs_embeds is None:
|
||||
inputs_embeds = self.embed_tokens(input_ids)
|
||||
# embed positions
|
||||
if attention_mask is None:
|
||||
attention_mask = torch.ones(
|
||||
(batch_size, seq_length_with_past), dtype=torch.bool, device=inputs_embeds.device
|
||||
)
|
||||
attention_mask = self._prepare_decoder_attention_mask(
|
||||
attention_mask, (batch_size, seq_length), inputs_embeds, past_key_values_length
|
||||
)
|
||||
|
||||
hidden_states = inputs_embeds
|
||||
|
||||
if self.gradient_checkpointing and self.training:
|
||||
if use_cache:
|
||||
logger.warning_once(
|
||||
"`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
|
||||
)
|
||||
use_cache = False
|
||||
|
||||
# decoder layers
|
||||
all_hidden_states = () if output_hidden_states else None
|
||||
all_self_attns = () if output_attentions else None
|
||||
next_decoder_cache = () if use_cache else None
|
||||
|
||||
for idx, decoder_layer in enumerate(self.layers):
|
||||
if output_hidden_states:
|
||||
all_hidden_states += (hidden_states,)
|
||||
|
||||
past_key_value = past_key_values[idx] if past_key_values is not None else None
|
||||
|
||||
if self.gradient_checkpointing and self.training:
|
||||
|
||||
def create_custom_forward(module):
|
||||
def custom_forward(*inputs):
|
||||
# None for past_key_value
|
||||
return module(*inputs, output_attentions, None)
|
||||
|
||||
return custom_forward
|
||||
|
||||
layer_outputs = torch.utils.checkpoint.checkpoint(
|
||||
create_custom_forward(decoder_layer),
|
||||
hidden_states,
|
||||
attention_mask,
|
||||
position_ids,
|
||||
None,
|
||||
)
|
||||
else:
|
||||
layer_outputs = decoder_layer(
|
||||
hidden_states,
|
||||
attention_mask=attention_mask,
|
||||
position_ids=position_ids,
|
||||
past_key_value=past_key_value,
|
||||
output_attentions=output_attentions,
|
||||
use_cache=use_cache,
|
||||
)
|
||||
|
||||
hidden_states = layer_outputs[0]
|
||||
|
||||
if use_cache:
|
||||
next_decoder_cache += (layer_outputs[2 if output_attentions else 1],)
|
||||
|
||||
if output_attentions:
|
||||
all_self_attns += (layer_outputs[1],)
|
||||
|
||||
hidden_states = self.norm(hidden_states)
|
||||
|
||||
# add hidden states from the last decoder layer
|
||||
if output_hidden_states:
|
||||
all_hidden_states += (hidden_states,)
|
||||
|
||||
next_cache = next_decoder_cache if use_cache else None
|
||||
if not return_dict:
|
||||
return tuple(v for v in [hidden_states, next_cache, all_hidden_states, all_self_attns] if v is not None)
|
||||
return BaseModelOutputWithPast(
|
||||
last_hidden_state=hidden_states,
|
||||
past_key_values=next_cache,
|
||||
hidden_states=all_hidden_states,
|
||||
attentions=all_self_attns,
|
||||
)
|
||||
|
||||
|
||||
class InternLMForCausalLM(InternLMPreTrainedModel):
|
||||
_auto_class = "AutoModelForCausalLM"
|
||||
|
||||
def __init__(self, config):
|
||||
super().__init__(config)
|
||||
self.model = InternLMModel(config)
|
||||
|
||||
self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
|
||||
|
||||
# Initialize weights and apply final processing
|
||||
self.post_init()
|
||||
|
||||
def get_input_embeddings(self):
|
||||
return self.model.embed_tokens
|
||||
|
||||
def set_input_embeddings(self, value):
|
||||
self.model.embed_tokens = value
|
||||
|
||||
def get_output_embeddings(self):
|
||||
return self.lm_head
|
||||
|
||||
def set_output_embeddings(self, new_embeddings):
|
||||
self.lm_head = new_embeddings
|
||||
|
||||
def set_decoder(self, decoder):
|
||||
self.model = decoder
|
||||
|
||||
def get_decoder(self):
|
||||
return self.model
|
||||
|
||||
@add_start_docstrings_to_model_forward(INTERNLM_INPUTS_DOCSTRING)
|
||||
@replace_return_docstrings(output_type=CausalLMOutputWithPast, config_class=_CONFIG_FOR_DOC)
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.LongTensor = None,
|
||||
attention_mask: Optional[torch.Tensor] = None,
|
||||
position_ids: Optional[torch.LongTensor] = None,
|
||||
past_key_values: Optional[List[torch.FloatTensor]] = None,
|
||||
inputs_embeds: Optional[torch.FloatTensor] = None,
|
||||
labels: Optional[torch.LongTensor] = None,
|
||||
use_cache: Optional[bool] = None,
|
||||
output_attentions: Optional[bool] = None,
|
||||
output_hidden_states: Optional[bool] = None,
|
||||
return_dict: Optional[bool] = None,
|
||||
) -> Union[Tuple, CausalLMOutputWithPast]:
|
||||
r"""
|
||||
Args:
|
||||
labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
|
||||
Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
|
||||
config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
|
||||
(masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
|
||||
|
||||
Returns:
|
||||
|
||||
Example:
|
||||
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, InternLMForCausalLM
|
||||
|
||||
>>> model = InternLMForCausalLM.from_pretrained(PATH_TO_CONVERTED_WEIGHTS)
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained(PATH_TO_CONVERTED_TOKENIZER)
|
||||
|
||||
>>> prompt = "Hey, are you consciours? Can you talk to me?"
|
||||
>>> inputs = tokenizer(prompt, return_tensors="pt")
|
||||
|
||||
>>> # Generate
|
||||
>>> generate_ids = model.generate(inputs.input_ids, max_length=30)
|
||||
>>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
|
||||
"Hey, are you consciours? Can you talk to me?\nI'm not consciours, but I can talk to you."
|
||||
```"""
|
||||
|
||||
output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
|
||||
output_hidden_states = (
|
||||
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
|
||||
)
|
||||
return_dict = return_dict if return_dict is not None else self.config.use_return_dict
|
||||
|
||||
# decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
|
||||
outputs = self.model(
|
||||
input_ids=input_ids,
|
||||
attention_mask=attention_mask,
|
||||
position_ids=position_ids,
|
||||
past_key_values=past_key_values,
|
||||
inputs_embeds=inputs_embeds,
|
||||
use_cache=use_cache,
|
||||
output_attentions=output_attentions,
|
||||
output_hidden_states=output_hidden_states,
|
||||
return_dict=return_dict,
|
||||
)
|
||||
|
||||
hidden_states = outputs[0]
|
||||
logits = self.lm_head(hidden_states)
|
||||
|
||||
loss = None
|
||||
if labels is not None:
|
||||
# Shift so that tokens < n predict n
|
||||
shift_logits = logits[..., :-1, :].contiguous()
|
||||
shift_labels = labels[..., 1:].contiguous()
|
||||
# Flatten the tokens
|
||||
loss_fct = CrossEntropyLoss()
|
||||
shift_logits = shift_logits.view(-1, self.config.vocab_size)
|
||||
shift_labels = shift_labels.view(-1)
|
||||
# Enable model parallelism
|
||||
shift_labels = shift_labels.to(shift_logits.device)
|
||||
loss = loss_fct(shift_logits, shift_labels)
|
||||
|
||||
if not return_dict:
|
||||
output = (logits,) + outputs[1:]
|
||||
return (loss,) + output if loss is not None else output
|
||||
|
||||
return CausalLMOutputWithPast(
|
||||
loss=loss,
|
||||
logits=logits,
|
||||
past_key_values=outputs.past_key_values,
|
||||
hidden_states=outputs.hidden_states,
|
||||
attentions=outputs.attentions,
|
||||
)
|
||||
|
||||
def prepare_inputs_for_generation(
|
||||
self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, **kwargs
|
||||
):
|
||||
if past_key_values:
|
||||
input_ids = input_ids[:, -1:]
|
||||
|
||||
position_ids = kwargs.get("position_ids", None)
|
||||
if attention_mask is not None and position_ids is None:
|
||||
# create position_ids on the fly for batch generation
|
||||
position_ids = attention_mask.long().cumsum(-1) - 1
|
||||
position_ids.masked_fill_(attention_mask == 0, 1)
|
||||
if past_key_values:
|
||||
position_ids = position_ids[:, -1].unsqueeze(-1)
|
||||
|
||||
# if `inputs_embeds` are passed, we only want to use them in the 1st generation step
|
||||
if inputs_embeds is not None and past_key_values is None:
|
||||
model_inputs = {"inputs_embeds": inputs_embeds}
|
||||
else:
|
||||
model_inputs = {"input_ids": input_ids}
|
||||
|
||||
model_inputs.update(
|
||||
{
|
||||
"position_ids": position_ids,
|
||||
"past_key_values": past_key_values,
|
||||
"use_cache": kwargs.get("use_cache"),
|
||||
"attention_mask": attention_mask,
|
||||
}
|
||||
)
|
||||
return model_inputs
|
||||
|
||||
@staticmethod
|
||||
def _reorder_cache(past_key_values, beam_idx):
|
||||
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
|
||||
|
||||
def build_inputs(self, tokenizer, query: str, history: List[Tuple[str, str]] = []):
|
||||
prompt = ""
|
||||
for record in history:
|
||||
prompt += f"""<|User|>:{record[0]}<eoh>\n<|Bot|>:{record[1]}<eoa>\n"""
|
||||
prompt += f"""<|User|>:{query}<eoh>\n<|Bot|>:"""
|
||||
return tokenizer([prompt], return_tensors="pt")
|
||||
|
||||
@torch.no_grad()
|
||||
def chat(self,
|
||||
tokenizer,
|
||||
query: str,
|
||||
history: List[Tuple[str, str]] = [],
|
||||
streamer: Optional[BaseStreamer] = None,
|
||||
max_new_tokens: int = 1024,
|
||||
do_sample: bool = True,
|
||||
temperature: float = 0.8,
|
||||
top_p: float = 0.8,
|
||||
**kwargs):
|
||||
inputs = self.build_inputs(tokenizer, query, history)
|
||||
inputs = {k: v.to(self.device) for k, v in inputs.items() if torch.is_tensor(v)}
|
||||
outputs = self.generate(**inputs,
|
||||
streamer=streamer,
|
||||
max_new_tokens=max_new_tokens,
|
||||
do_sample=do_sample,
|
||||
temperature=temperature,
|
||||
top_p=top_p,
|
||||
**kwargs)
|
||||
outputs = outputs[0].cpu().tolist()[len(inputs["input_ids"][0]):]
|
||||
response = tokenizer.decode(outputs, skip_special_tokens=True)
|
||||
response = response.split("<eoa>")[0]
|
||||
history = history + [(query, response)]
|
||||
return response, history
|
||||
|
||||
@torch.no_grad()
|
||||
def stream_chat(self,
|
||||
tokenizer,
|
||||
query: str,
|
||||
history: List[Tuple[str, str]] = [],
|
||||
max_new_tokens: int = 1024,
|
||||
do_sample: bool = True,
|
||||
temperature: float = 0.8,
|
||||
top_p: float = 0.8,
|
||||
**kwargs):
|
||||
"""
|
||||
Return a generator in format: (response, history)
|
||||
Eg.
|
||||
('你好，有什么可以帮助您的吗', [('你好', '你好，有什么可以帮助您的吗')])
|
||||
('你好，有什么可以帮助您的吗？', [('你好', '你好，有什么可以帮助您的吗？')])
|
||||
"""
|
||||
|
||||
response_queue = queue.Queue(maxsize=20)
|
||||
|
||||
class ChatStreamer(BaseStreamer):
|
||||
def __init__(self, tokenizer) -> None:
|
||||
super().__init__()
|
||||
self.tokenizer = tokenizer
|
||||
self.queue = response_queue
|
||||
self.query = query
|
||||
self.history = history
|
||||
self.response = ""
|
||||
self.received_inputs = False
|
||||
self.queue.put((self.response, history + [(self.query, self.response)]))
|
||||
|
||||
def put(self, value):
|
||||
if len(value.shape) > 1 and value.shape[0] > 1:
|
||||
raise ValueError("ChatStreamer only supports batch size 1")
|
||||
elif len(value.shape) > 1:
|
||||
value = value[0]
|
||||
|
||||
if not self.received_inputs:
|
||||
# The first received value is input_ids, ignore here
|
||||
self.received_inputs = True
|
||||
return
|
||||
|
||||
token = self.tokenizer.decode([value[-1]], skip_special_tokens=True)
|
||||
if token.strip() != "<eoa>":
|
||||
self.response = self.response + token
|
||||
history = self.history + [(self.query, self.response)]
|
||||
self.queue.put((self.response, history))
|
||||
|
||||
def end(self):
|
||||
self.queue.put(None)
|
||||
|
||||
def stream_producer():
|
||||
return self.chat(
|
||||
tokenizer=tokenizer,
|
||||
query=query,
|
||||
streamer=ChatStreamer(tokenizer=tokenizer),
|
||||
history=history,
|
||||
max_new_tokens=max_new_tokens,
|
||||
do_sample=do_sample,
|
||||
temperature=temperature,
|
||||
top_p=top_p,
|
||||
**kwargs
|
||||
)
|
||||
|
||||
def consumer():
|
||||
producer = threading.Thread(target=stream_producer)
|
||||
producer.start()
|
||||
while True:
|
||||
res = response_queue.get()
|
||||
if res is None:
|
||||
return
|
||||
yield res
|
||||
|
||||
return consumer()
|
||||
|
||||
|
||||
@add_start_docstrings(
|
||||
"""
|
||||
The InternLM Model transformer with a sequence classification head on top (linear layer).
|
||||
|
||||
[`InternLMForSequenceClassification`] uses the last token in order to do the classification, as other causal models
|
||||
(e.g. GPT-2) do.
|
||||
|
||||
Since it does classification on the last token, it requires to know the position of the last token. If a
|
||||
`pad_token_id` is defined in the configuration, it finds the last token that is not a padding token in each row. If
|
||||
no `pad_token_id` is defined, it simply takes the last value in each row of the batch. Since it cannot guess the
|
||||
padding tokens when `inputs_embeds` are passed instead of `input_ids`, it does the same (take the last value in
|
||||
each row of the batch).
|
||||
""",
|
||||
INTERNLM_START_DOCSTRING,
|
||||
)
|
||||
class InternLMForSequenceClassification(InternLMPreTrainedModel):
|
||||
_keys_to_ignore_on_load_missing = [r"lm_head.weight"]
|
||||
|
||||
def __init__(self, config):
|
||||
super().__init__(config)
|
||||
self.num_labels = config.num_labels
|
||||
self.model = InternLMModel(config)
|
||||
self.score = nn.Linear(config.hidden_size, self.num_labels, bias=False)
|
||||
|
||||
# Initialize weights and apply final processing
|
||||
self.post_init()
|
||||
|
||||
def get_input_embeddings(self):
|
||||
return self.model.embed_tokens
|
||||
|
||||
def set_input_embeddings(self, value):
|
||||
self.model.embed_tokens = value
|
||||
|
||||
@add_start_docstrings_to_model_forward(INTERNLM_INPUTS_DOCSTRING)
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.LongTensor = None,
|
||||
attention_mask: Optional[torch.Tensor] = None,
|
||||
position_ids: Optional[torch.LongTensor] = None,
|
||||
past_key_values: Optional[List[torch.FloatTensor]] = None,
|
||||
inputs_embeds: Optional[torch.FloatTensor] = None,
|
||||
labels: Optional[torch.LongTensor] = None,
|
||||
use_cache: Optional[bool] = None,
|
||||
output_attentions: Optional[bool] = None,
|
||||
output_hidden_states: Optional[bool] = None,
|
||||
return_dict: Optional[bool] = None,
|
||||
) -> Union[Tuple, SequenceClassifierOutputWithPast]:
|
||||
r"""
|
||||
labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
|
||||
Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
|
||||
config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
|
||||
`config.num_labels > 1` a classification loss is computed (Cross-Entropy).
|
||||
"""
|
||||
return_dict = return_dict if return_dict is not None else self.config.use_return_dict
|
||||
|
||||
transformer_outputs = self.model(
|
||||
input_ids,
|
||||
attention_mask=attention_mask,
|
||||
position_ids=position_ids,
|
||||
past_key_values=past_key_values,
|
||||
inputs_embeds=inputs_embeds,
|
||||
use_cache=use_cache,
|
||||
output_attentions=output_attentions,
|
||||
output_hidden_states=output_hidden_states,
|
||||
return_dict=return_dict,
|
||||
)
|
||||
hidden_states = transformer_outputs[0]
|
||||
logits = self.score(hidden_states)
|
||||
|
||||
if input_ids is not None:
|
||||
batch_size = input_ids.shape[0]
|
||||
else:
|
||||
batch_size = inputs_embeds.shape[0]
|
||||
|
||||
if self.config.pad_token_id is None and batch_size != 1:
|
||||
raise ValueError("Cannot handle batch sizes > 1 if no padding token is defined.")
|
||||
if self.config.pad_token_id is None:
|
||||
sequence_lengths = -1
|
||||
else:
|
||||
if input_ids is not None:
|
||||
sequence_lengths = (torch.ne(input_ids, self.config.pad_token_id).sum(-1) - 1).to(logits.device)
|
||||
else:
|
||||
sequence_lengths = -1
|
||||
|
||||
pooled_logits = logits[torch.arange(batch_size, device=logits.device), sequence_lengths]
|
||||
|
||||
loss = None
|
||||
if labels is not None:
|
||||
labels = labels.to(logits.device)
|
||||
if self.config.problem_type is None:
|
||||
if self.num_labels == 1:
|
||||
self.config.problem_type = "regression"
|
||||
elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
|
||||
self.config.problem_type = "single_label_classification"
|
||||
else:
|
||||
self.config.problem_type = "multi_label_classification"
|
||||
|
||||
if self.config.problem_type == "regression":
|
||||
loss_fct = MSELoss()
|
||||
if self.num_labels == 1:
|
||||
loss = loss_fct(pooled_logits.squeeze(), labels.squeeze())
|
||||
else:
|
||||
loss = loss_fct(pooled_logits, labels)
|
||||
elif self.config.problem_type == "single_label_classification":
|
||||
loss_fct = CrossEntropyLoss()
|
||||
loss = loss_fct(pooled_logits.view(-1, self.num_labels), labels.view(-1))
|
||||
elif self.config.problem_type == "multi_label_classification":
|
||||
loss_fct = BCEWithLogitsLoss()
|
||||
loss = loss_fct(pooled_logits, labels)
|
||||
if not return_dict:
|
||||
output = (pooled_logits,) + transformer_outputs[1:]
|
||||
return ((loss,) + output) if loss is not None else output
|
||||
|
||||
return SequenceClassifierOutputWithPast(
|
||||
loss=loss,
|
||||
logits=pooled_logits,
|
||||
past_key_values=transformer_outputs.past_key_values,
|
||||
hidden_states=transformer_outputs.hidden_states,
|
||||
attentions=transformer_outputs.attentions,
|
||||
)
|
||||
3
pytorch_model-00001-of-00016.bin
Normal file
3
pytorch_model-00001-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:7292ecf6ca2ff8c357339835b054391eec08f3abb5f88b876151dda29ddc5efa
|
||||
size 979406258
|
||||
3
pytorch_model-00002-of-00016.bin
Normal file
3
pytorch_model-00002-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:557499dba8a48f08602c6311643f15b57ae8ec829a73fe5fa95d79c28bc3656d
|
||||
size 989963655
|
||||
3
pytorch_model-00003-of-00016.bin
Normal file
3
pytorch_model-00003-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:9864bb73b7b06cb684857f645f100b7eb4d5bcc3989dac0ba5ce95fd0ef430d1
|
||||
size 966929505
|
||||
3
pytorch_model-00004-of-00016.bin
Normal file
3
pytorch_model-00004-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:c7b026ad848ae614c7863dafe25988c07d35d0619c82051b6c8d73ea5b9c4c3a
|
||||
size 966912445
|
||||
3
pytorch_model-00005-of-00016.bin
Normal file
3
pytorch_model-00005-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:45f4352660697dfca454556ae9ac4c6511377b0f8046019b704f9760b2fd52ae
|
||||
size 989980715
|
||||
3
pytorch_model-00006-of-00016.bin
Normal file
3
pytorch_model-00006-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:c119f0844a55176a178de427c3a2d349fc050ea7b6190df1c4376a4669874e2a
|
||||
size 943861171
|
||||
3
pytorch_model-00007-of-00016.bin
Normal file
3
pytorch_model-00007-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:1ca9a5f374ae4025e48fcdef9007be505ea056fa56bc220756faebbcebaeda2c
|
||||
size 989963719
|
||||
3
pytorch_model-00008-of-00016.bin
Normal file
3
pytorch_model-00008-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:3a217aa39bf8ed33eef29d7c96d60db88941492a4394e2cdd0f383fbbf602075
|
||||
size 966929505
|
||||
3
pytorch_model-00009-of-00016.bin
Normal file
3
pytorch_model-00009-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:e30f124b49ec6ad6054659fea786759995cbc6a7066d192b92160c95bd5e5414
|
||||
size 966912445
|
||||
3
pytorch_model-00010-of-00016.bin
Normal file
3
pytorch_model-00010-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:a4e3bf22d4cdba303857bbd560846dc102a2436b7a81210ad5475e193d0c2fbf
|
||||
size 989980779
|
||||
3
pytorch_model-00011-of-00016.bin
Normal file
3
pytorch_model-00011-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:0d018139dbd3b1b57b79d0e7d2761fe21bc876087e1c83876f081a9d99366dbc
|
||||
size 943861171
|
||||
3
pytorch_model-00012-of-00016.bin
Normal file
3
pytorch_model-00012-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:a5e2ad0d87adec50f0190cd24631908950c5de602322ab4855da18842a5de618
|
||||
size 989963719
|
||||
3
pytorch_model-00013-of-00016.bin
Normal file
3
pytorch_model-00013-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:52350977c882f29e2b8060096e0b991d2ba003bfb2f18acaa03a2162063ac7b8
|
||||
size 966929505
|
||||
3
pytorch_model-00014-of-00016.bin
Normal file
3
pytorch_model-00014-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:2e74013e34bcf7484138fc5e37c7e30f63d7246abde8d57a90daf5e4f8572152
|
||||
size 966912445
|
||||
3
pytorch_model-00015-of-00016.bin
Normal file
3
pytorch_model-00015-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:5208f2fad83ac11b1af34bca80630cb37a701ad272104819e2039eb3a7b01f34
|
||||
size 180381870
|
||||
3
pytorch_model-00016-of-00016.bin
Normal file
3
pytorch_model-00016-of-00016.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:c96a26a999a0034cf7fb5c1d0b9c8185cca71acf9f504f5e7fd482be7c10bf4f
|
||||
size 845153194
|
||||
426
pytorch_model.bin.index.json
Normal file
426
pytorch_model.bin.index.json
Normal file
@@ -0,0 +1,426 @@
|
||||
{
|
||||
"metadata": {
|
||||
"total_size": 14643896320
|
||||
},
|
||||
"weight_map": {
|
||||
"lm_head.weight": "pytorch_model-00016-of-00016.bin",
|
||||
"model.embed_tokens.weight": "pytorch_model-00001-of-00016.bin",
|
||||
"model.layers.0.input_layernorm.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.0.mlp.down_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.0.mlp.gate_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.0.mlp.up_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.0.post_attention_layernorm.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.0.self_attn.k_proj.bias": "pytorch_model-00001-of-00016.bin",
|
||||
"model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00016.bin",
|
||||
"model.layers.0.self_attn.o_proj.bias": "pytorch_model-00001-of-00016.bin",
|
||||
"model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00016.bin",
|
||||
"model.layers.0.self_attn.q_proj.bias": "pytorch_model-00001-of-00016.bin",
|
||||
"model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00016.bin",
|
||||
"model.layers.0.self_attn.v_proj.bias": "pytorch_model-00001-of-00016.bin",
|
||||
"model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00016.bin",
|
||||
"model.layers.1.input_layernorm.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.mlp.down_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.mlp.gate_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.mlp.up_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.post_attention_layernorm.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.self_attn.k_proj.bias": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.self_attn.k_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.self_attn.o_proj.bias": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.self_attn.o_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.self_attn.q_proj.bias": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.self_attn.q_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.self_attn.v_proj.bias": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.1.self_attn.v_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.10.input_layernorm.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.mlp.down_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.mlp.gate_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.mlp.up_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.post_attention_layernorm.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.self_attn.k_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.self_attn.k_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.self_attn.o_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.self_attn.o_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.self_attn.q_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.self_attn.q_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.self_attn.v_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.10.self_attn.v_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.input_layernorm.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.mlp.down_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.mlp.gate_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.mlp.up_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.post_attention_layernorm.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.self_attn.k_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.self_attn.k_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.self_attn.o_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.self_attn.o_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.self_attn.q_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.self_attn.q_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.self_attn.v_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.11.self_attn.v_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.12.input_layernorm.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.12.mlp.down_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.12.mlp.gate_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.12.mlp.up_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.12.post_attention_layernorm.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.12.self_attn.k_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.12.self_attn.k_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.12.self_attn.o_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.12.self_attn.o_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.12.self_attn.q_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.12.self_attn.q_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.12.self_attn.v_proj.bias": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.12.self_attn.v_proj.weight": "pytorch_model-00006-of-00016.bin",
|
||||
"model.layers.13.input_layernorm.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.mlp.down_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.mlp.gate_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.mlp.up_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.post_attention_layernorm.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.self_attn.k_proj.bias": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.self_attn.k_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.self_attn.o_proj.bias": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.self_attn.o_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.self_attn.q_proj.bias": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.self_attn.q_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.self_attn.v_proj.bias": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.13.self_attn.v_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.14.input_layernorm.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.14.mlp.down_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.14.mlp.gate_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.14.mlp.up_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.14.post_attention_layernorm.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.14.self_attn.k_proj.bias": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.14.self_attn.k_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.14.self_attn.o_proj.bias": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.14.self_attn.o_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.14.self_attn.q_proj.bias": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.14.self_attn.q_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.14.self_attn.v_proj.bias": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.14.self_attn.v_proj.weight": "pytorch_model-00007-of-00016.bin",
|
||||
"model.layers.15.input_layernorm.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.mlp.down_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.mlp.gate_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.mlp.up_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.post_attention_layernorm.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.self_attn.k_proj.bias": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.self_attn.k_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.self_attn.o_proj.bias": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.self_attn.o_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.self_attn.q_proj.bias": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.self_attn.q_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.self_attn.v_proj.bias": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.15.self_attn.v_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.input_layernorm.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.mlp.down_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.mlp.gate_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.mlp.up_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.post_attention_layernorm.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.self_attn.k_proj.bias": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.self_attn.k_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.self_attn.o_proj.bias": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.self_attn.o_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.self_attn.q_proj.bias": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.self_attn.q_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.self_attn.v_proj.bias": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.16.self_attn.v_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.17.input_layernorm.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.17.mlp.down_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.17.mlp.gate_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.17.mlp.up_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.17.post_attention_layernorm.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.17.self_attn.k_proj.bias": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.17.self_attn.k_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.17.self_attn.o_proj.bias": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.17.self_attn.o_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.17.self_attn.q_proj.bias": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.17.self_attn.q_proj.weight": "pytorch_model-00008-of-00016.bin",
|
||||
"model.layers.17.self_attn.v_proj.bias": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.17.self_attn.v_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.input_layernorm.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.mlp.down_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.mlp.gate_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.mlp.up_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.post_attention_layernorm.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.self_attn.k_proj.bias": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.self_attn.k_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.self_attn.o_proj.bias": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.self_attn.o_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.self_attn.q_proj.bias": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.self_attn.q_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.self_attn.v_proj.bias": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.18.self_attn.v_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.19.input_layernorm.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.19.mlp.down_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.19.mlp.gate_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.19.mlp.up_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.19.post_attention_layernorm.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.19.self_attn.k_proj.bias": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.19.self_attn.k_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.19.self_attn.o_proj.bias": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.19.self_attn.o_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.19.self_attn.q_proj.bias": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.19.self_attn.q_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.19.self_attn.v_proj.bias": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.19.self_attn.v_proj.weight": "pytorch_model-00009-of-00016.bin",
|
||||
"model.layers.2.input_layernorm.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.2.mlp.down_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.2.mlp.gate_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.2.mlp.up_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.2.post_attention_layernorm.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.2.self_attn.k_proj.bias": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.2.self_attn.k_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.2.self_attn.o_proj.bias": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.2.self_attn.o_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.2.self_attn.q_proj.bias": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.2.self_attn.q_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.2.self_attn.v_proj.bias": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.2.self_attn.v_proj.weight": "pytorch_model-00002-of-00016.bin",
|
||||
"model.layers.20.input_layernorm.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.mlp.down_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.mlp.gate_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.mlp.up_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.post_attention_layernorm.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.self_attn.k_proj.bias": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.self_attn.k_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.self_attn.o_proj.bias": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.self_attn.o_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.self_attn.q_proj.bias": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.self_attn.q_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.self_attn.v_proj.bias": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.20.self_attn.v_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.input_layernorm.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.mlp.down_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.mlp.gate_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.mlp.up_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.post_attention_layernorm.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.self_attn.k_proj.bias": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.self_attn.k_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.self_attn.o_proj.bias": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.self_attn.o_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.self_attn.q_proj.bias": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.self_attn.q_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.self_attn.v_proj.bias": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.21.self_attn.v_proj.weight": "pytorch_model-00010-of-00016.bin",
|
||||
"model.layers.22.input_layernorm.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.mlp.down_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.mlp.gate_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.mlp.up_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.post_attention_layernorm.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.self_attn.k_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.self_attn.k_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.self_attn.o_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.self_attn.o_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.self_attn.q_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.self_attn.q_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.self_attn.v_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.22.self_attn.v_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.input_layernorm.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.mlp.down_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.mlp.gate_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.mlp.up_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.post_attention_layernorm.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.self_attn.k_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.self_attn.k_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.self_attn.o_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.self_attn.o_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.self_attn.q_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.self_attn.q_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.self_attn.v_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.23.self_attn.v_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.24.input_layernorm.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.24.mlp.down_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.24.mlp.gate_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.24.mlp.up_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.24.post_attention_layernorm.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.24.self_attn.k_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.24.self_attn.k_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.24.self_attn.o_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.24.self_attn.o_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.24.self_attn.q_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.24.self_attn.q_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.24.self_attn.v_proj.bias": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.24.self_attn.v_proj.weight": "pytorch_model-00011-of-00016.bin",
|
||||
"model.layers.25.input_layernorm.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.mlp.down_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.mlp.gate_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.mlp.up_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.post_attention_layernorm.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.self_attn.k_proj.bias": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.self_attn.k_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.self_attn.o_proj.bias": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.self_attn.o_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.self_attn.q_proj.bias": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.self_attn.q_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.self_attn.v_proj.bias": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.25.self_attn.v_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.26.input_layernorm.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.26.mlp.down_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.26.mlp.gate_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.26.mlp.up_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.26.post_attention_layernorm.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.26.self_attn.k_proj.bias": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.26.self_attn.k_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.26.self_attn.o_proj.bias": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.26.self_attn.o_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.26.self_attn.q_proj.bias": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.26.self_attn.q_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.26.self_attn.v_proj.bias": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.26.self_attn.v_proj.weight": "pytorch_model-00012-of-00016.bin",
|
||||
"model.layers.27.input_layernorm.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.mlp.down_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.mlp.gate_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.mlp.up_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.post_attention_layernorm.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.self_attn.k_proj.bias": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.self_attn.k_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.self_attn.o_proj.bias": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.self_attn.o_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.self_attn.q_proj.bias": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.self_attn.q_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.self_attn.v_proj.bias": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.27.self_attn.v_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.input_layernorm.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.mlp.down_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.mlp.gate_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.mlp.up_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.post_attention_layernorm.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.self_attn.k_proj.bias": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.self_attn.k_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.self_attn.o_proj.bias": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.self_attn.o_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.self_attn.q_proj.bias": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.self_attn.q_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.self_attn.v_proj.bias": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.28.self_attn.v_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.29.input_layernorm.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.29.mlp.down_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.29.mlp.gate_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.29.mlp.up_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.29.post_attention_layernorm.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.29.self_attn.k_proj.bias": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.29.self_attn.k_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.29.self_attn.o_proj.bias": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.29.self_attn.o_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.29.self_attn.q_proj.bias": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.29.self_attn.q_proj.weight": "pytorch_model-00013-of-00016.bin",
|
||||
"model.layers.29.self_attn.v_proj.bias": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.29.self_attn.v_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.3.input_layernorm.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.mlp.down_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.mlp.gate_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.mlp.up_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.post_attention_layernorm.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.self_attn.k_proj.bias": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.self_attn.k_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.self_attn.o_proj.bias": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.self_attn.o_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.self_attn.q_proj.bias": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.self_attn.q_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.self_attn.v_proj.bias": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.3.self_attn.v_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.30.input_layernorm.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.mlp.down_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.mlp.gate_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.mlp.up_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.post_attention_layernorm.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.self_attn.k_proj.bias": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.self_attn.k_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.self_attn.o_proj.bias": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.self_attn.o_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.self_attn.q_proj.bias": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.self_attn.q_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.self_attn.v_proj.bias": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.30.self_attn.v_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.31.input_layernorm.weight": "pytorch_model-00015-of-00016.bin",
|
||||
"model.layers.31.mlp.down_proj.weight": "pytorch_model-00015-of-00016.bin",
|
||||
"model.layers.31.mlp.gate_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.31.mlp.up_proj.weight": "pytorch_model-00015-of-00016.bin",
|
||||
"model.layers.31.post_attention_layernorm.weight": "pytorch_model-00015-of-00016.bin",
|
||||
"model.layers.31.self_attn.k_proj.bias": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.31.self_attn.k_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.31.self_attn.o_proj.bias": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.31.self_attn.o_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.31.self_attn.q_proj.bias": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.31.self_attn.q_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.31.self_attn.v_proj.bias": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.31.self_attn.v_proj.weight": "pytorch_model-00014-of-00016.bin",
|
||||
"model.layers.4.input_layernorm.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.mlp.down_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.mlp.gate_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.mlp.up_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.post_attention_layernorm.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.self_attn.k_proj.bias": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.self_attn.k_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.self_attn.o_proj.bias": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.self_attn.o_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.self_attn.q_proj.bias": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.self_attn.q_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.self_attn.v_proj.bias": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.4.self_attn.v_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.5.input_layernorm.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.5.mlp.down_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.5.mlp.gate_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.5.mlp.up_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.5.post_attention_layernorm.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.5.self_attn.k_proj.bias": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.5.self_attn.k_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.5.self_attn.o_proj.bias": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.5.self_attn.o_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.5.self_attn.q_proj.bias": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.5.self_attn.q_proj.weight": "pytorch_model-00003-of-00016.bin",
|
||||
"model.layers.5.self_attn.v_proj.bias": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.5.self_attn.v_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.input_layernorm.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.mlp.down_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.mlp.gate_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.mlp.up_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.post_attention_layernorm.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.self_attn.k_proj.bias": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.self_attn.k_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.self_attn.o_proj.bias": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.self_attn.o_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.self_attn.q_proj.bias": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.self_attn.q_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.self_attn.v_proj.bias": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.6.self_attn.v_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.7.input_layernorm.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.7.mlp.down_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.7.mlp.gate_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.7.mlp.up_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.7.post_attention_layernorm.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.7.self_attn.k_proj.bias": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.7.self_attn.k_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.7.self_attn.o_proj.bias": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.7.self_attn.o_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.7.self_attn.q_proj.bias": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.7.self_attn.q_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.7.self_attn.v_proj.bias": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.7.self_attn.v_proj.weight": "pytorch_model-00004-of-00016.bin",
|
||||
"model.layers.8.input_layernorm.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.mlp.down_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.mlp.gate_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.mlp.up_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.post_attention_layernorm.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.self_attn.k_proj.bias": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.self_attn.k_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.self_attn.o_proj.bias": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.self_attn.o_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.self_attn.q_proj.bias": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.self_attn.q_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.self_attn.v_proj.bias": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.8.self_attn.v_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.input_layernorm.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.mlp.down_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.mlp.gate_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.mlp.up_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.post_attention_layernorm.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.self_attn.k_proj.bias": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.self_attn.k_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.self_attn.o_proj.bias": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.self_attn.o_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.self_attn.q_proj.bias": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.self_attn.q_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.self_attn.v_proj.bias": "pytorch_model-00005-of-00016.bin",
|
||||
"model.layers.9.self_attn.v_proj.weight": "pytorch_model-00005-of-00016.bin",
|
||||
"model.norm.weight": "pytorch_model-00015-of-00016.bin"
|
||||
}
|
||||
}
|
||||
6
special_tokens_map.json
Normal file
6
special_tokens_map.json
Normal file
@@ -0,0 +1,6 @@
|
||||
{
|
||||
"bos_token": "<s>",
|
||||
"eos_token": "</s>",
|
||||
"pad_token": "</s>",
|
||||
"unk_token": "<unk>"
|
||||
}
|
||||
241
tokenization_internlm.py
Normal file
241
tokenization_internlm.py
Normal file
@@ -0,0 +1,241 @@
|
||||
# coding=utf-8
|
||||
# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
|
||||
#
|
||||
# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
|
||||
# and OPT implementations in this library. It has been modified from its
|
||||
# original forms to accommodate minor architectural differences compared
|
||||
# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
"""Tokenization classes for IntermLM."""
|
||||
import os
|
||||
from shutil import copyfile
|
||||
from typing import Any, Dict, List, Optional, Tuple
|
||||
|
||||
import sentencepiece as spm
|
||||
|
||||
from transformers.tokenization_utils import PreTrainedTokenizer
|
||||
from transformers.utils import logging
|
||||
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
VOCAB_FILES_NAMES = {"vocab_file": "./tokenizer.model"}
|
||||
|
||||
PRETRAINED_VOCAB_FILES_MAP = {}
|
||||
|
||||
|
||||
class InternLMTokenizer(PreTrainedTokenizer):
|
||||
"""
|
||||
Construct a InternLM tokenizer. Based on byte-level Byte-Pair-Encoding.
|
||||
|
||||
Args:
|
||||
vocab_file (`str`):
|
||||
Path to the vocabulary file.
|
||||
"""
|
||||
|
||||
vocab_files_names = VOCAB_FILES_NAMES
|
||||
pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
|
||||
model_input_names = ["input_ids", "attention_mask"]
|
||||
_auto_class = "AutoTokenizer"
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
vocab_file,
|
||||
unk_token="<unk>",
|
||||
bos_token="<s>",
|
||||
eos_token="</s>",
|
||||
pad_token="</s>",
|
||||
sp_model_kwargs: Optional[Dict[str, Any]] = None,
|
||||
add_bos_token=True,
|
||||
add_eos_token=False,
|
||||
decode_with_prefix_space=False,
|
||||
clean_up_tokenization_spaces=False,
|
||||
**kwargs,
|
||||
):
|
||||
self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
|
||||
self.vocab_file = vocab_file
|
||||
self.add_bos_token = add_bos_token
|
||||
self.add_eos_token = add_eos_token
|
||||
self.decode_with_prefix_space = decode_with_prefix_space
|
||||
self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
|
||||
self.sp_model.Load(vocab_file)
|
||||
self._no_prefix_space_tokens = None
|
||||
super().__init__(
|
||||
bos_token=bos_token,
|
||||
eos_token=eos_token,
|
||||
unk_token=unk_token,
|
||||
pad_token=pad_token,
|
||||
clean_up_tokenization_spaces=clean_up_tokenization_spaces,
|
||||
**kwargs,
|
||||
)
|
||||
""" Initialisation"""
|
||||
|
||||
@property
|
||||
def no_prefix_space_tokens(self):
|
||||
if self._no_prefix_space_tokens is None:
|
||||
vocab = self.convert_ids_to_tokens(list(range(self.vocab_size)))
|
||||
self._no_prefix_space_tokens = {i for i, tok in enumerate(vocab) if not tok.startswith("▁")}
|
||||
return self._no_prefix_space_tokens
|
||||
|
||||
@property
|
||||
def vocab_size(self):
|
||||
"""Returns vocab size"""
|
||||
return self.sp_model.get_piece_size()
|
||||
|
||||
@property
|
||||
def bos_token_id(self) -> Optional[int]:
|
||||
return self.sp_model.bos_id()
|
||||
|
||||
@property
|
||||
def eos_token_id(self) -> Optional[int]:
|
||||
return self.sp_model.eos_id()
|
||||
|
||||
def get_vocab(self):
|
||||
"""Returns vocab as a dict"""
|
||||
vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
|
||||
vocab.update(self.added_tokens_encoder)
|
||||
return vocab
|
||||
|
||||
def _tokenize(self, text):
|
||||
"""Returns a tokenized string."""
|
||||
return self.sp_model.encode(text, out_type=str)
|
||||
|
||||
def _convert_token_to_id(self, token):
|
||||
"""Converts a token (str) in an id using the vocab."""
|
||||
return self.sp_model.piece_to_id(token)
|
||||
|
||||
def _convert_id_to_token(self, index):
|
||||
"""Converts an index (integer) in a token (str) using the vocab."""
|
||||
token = self.sp_model.IdToPiece(index)
|
||||
return token
|
||||
|
||||
def _maybe_add_prefix_space(self, tokens, decoded):
|
||||
if tokens and tokens[0] not in self.no_prefix_space_tokens:
|
||||
return " " + decoded
|
||||
else:
|
||||
return decoded
|
||||
|
||||
def convert_tokens_to_string(self, tokens):
|
||||
"""Converts a sequence of tokens (string) in a single string."""
|
||||
current_sub_tokens = []
|
||||
out_string = ""
|
||||
prev_is_special = False
|
||||
for token in tokens:
|
||||
# make sure that special tokens are not decoded using sentencepiece model
|
||||
if token in self.all_special_tokens:
|
||||
if not prev_is_special:
|
||||
out_string += " "
|
||||
out_string += self.sp_model.decode(current_sub_tokens) + token
|
||||
prev_is_special = True
|
||||
current_sub_tokens = []
|
||||
else:
|
||||
current_sub_tokens.append(token)
|
||||
prev_is_special = False
|
||||
out_string += self.sp_model.decode(current_sub_tokens)
|
||||
out_string = self.clean_up_tokenization(out_string)
|
||||
out_string = self._maybe_add_prefix_space(tokens=tokens, decoded=out_string)
|
||||
return out_string[1:]
|
||||
|
||||
def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
|
||||
"""
|
||||
Save the vocabulary and special tokens file to a directory.
|
||||
|
||||
Args:
|
||||
save_directory (`str`):
|
||||
The directory in which to save the vocabulary.
|
||||
|
||||
Returns:
|
||||
`Tuple(str)`: Paths to the files saved.
|
||||
"""
|
||||
if not os.path.isdir(save_directory):
|
||||
logger.error(f"Vocabulary path ({save_directory}) should be a directory")
|
||||
return
|
||||
out_vocab_file = os.path.join(
|
||||
save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
|
||||
)
|
||||
|
||||
if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
|
||||
copyfile(self.vocab_file, out_vocab_file)
|
||||
elif not os.path.isfile(self.vocab_file):
|
||||
with open(out_vocab_file, "wb") as fi:
|
||||
content_spiece_model = self.sp_model.serialized_model_proto()
|
||||
fi.write(content_spiece_model)
|
||||
|
||||
return (out_vocab_file,)
|
||||
|
||||
def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
|
||||
if self.add_bos_token:
|
||||
bos_token_ids = [self.bos_token_id]
|
||||
else:
|
||||
bos_token_ids = []
|
||||
|
||||
output = bos_token_ids + token_ids_0
|
||||
|
||||
if token_ids_1 is not None:
|
||||
output = output + token_ids_1
|
||||
|
||||
if self.add_eos_token:
|
||||
output = output + [self.eos_token_id]
|
||||
|
||||
return output
|
||||
|
||||
def get_special_tokens_mask(
|
||||
self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
|
||||
) -> List[int]:
|
||||
"""
|
||||
Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
|
||||
special tokens using the tokenizer `prepare_for_model` method.
|
||||
|
||||
Args:
|
||||
token_ids_0 (`List[int]`):
|
||||
List of IDs.
|
||||
token_ids_1 (`List[int]`, *optional*):
|
||||
Optional second list of IDs for sequence pairs.
|
||||
already_has_special_tokens (`bool`, *optional*, defaults to `False`):
|
||||
Whether or not the token list is already formatted with special tokens for the model.
|
||||
|
||||
Returns:
|
||||
`List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
|
||||
"""
|
||||
if already_has_special_tokens:
|
||||
return super().get_special_tokens_mask(
|
||||
token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
|
||||
)
|
||||
|
||||
if token_ids_1 is None:
|
||||
return [1] + ([0] * len(token_ids_0)) + [1]
|
||||
return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
|
||||
|
||||
def create_token_type_ids_from_sequences(
|
||||
self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
|
||||
) -> List[int]:
|
||||
"""
|
||||
Create a mask from the two sequences passed to be used in a sequence-pair classification task. T5 does not make
|
||||
use of token type ids, therefore a list of zeros is returned.
|
||||
|
||||
Args:
|
||||
token_ids_0 (`List[int]`):
|
||||
List of IDs.
|
||||
token_ids_1 (`List[int]`, *optional*):
|
||||
Optional second list of IDs for sequence pairs.
|
||||
|
||||
Returns:
|
||||
`List[int]`: List of zeros.
|
||||
"""
|
||||
eos = [self.eos_token_id]
|
||||
|
||||
if token_ids_1 is None:
|
||||
return len(token_ids_0 + eos) * [0]
|
||||
return len(token_ids_0 + eos + token_ids_1 + eos) * [0]
|
||||
3
tokenizer.model
Normal file
3
tokenizer.model
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:aab622d98c98677a1a51f969e25765154487bf3e85c7819db105db2fcacba83f
|
||||
size 1658691
|
||||
43
tokenizer_config.json
Normal file
43
tokenizer_config.json
Normal file
@@ -0,0 +1,43 @@
|
||||
{
|
||||
"added_tokens_decoder": {
|
||||
"0": {
|
||||
"content": "<unk>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"1": {
|
||||
"content": "<s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"2": {
|
||||
"content": "</s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
}
|
||||
},
|
||||
"additional_special_tokens": [],
|
||||
"auto_map": {
|
||||
"AutoTokenizer": [
|
||||
"tokenization_internlm.InternLMTokenizer",
|
||||
null
|
||||
]
|
||||
},
|
||||
"bos_token": "<s>",
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "</s>",
|
||||
"model_max_length": 1000000000000000019884624838656,
|
||||
"pad_token": "</s>",
|
||||
"tokenizer_class": "InternLMTokenizer",
|
||||
"tokenizer_file": null,
|
||||
"unk_token": "<unk>"
|
||||
}
|
||||
Reference in New Issue
Block a user