sanbuphy/tianji-wish-7b
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: sanbuphy/tianji-wish-7b
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-09-27 10:49:12 +08:00
commit 0cd95acb4b
20 changed files with 2268 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
.gitattributes vendored Normal file
View File

@@ -0,0 +1,34 @@
*.7z filter=lfs diff=lfs merge=lfs -text
*.arrow filter=lfs diff=lfs merge=lfs -text
*.bin filter=lfs diff=lfs merge=lfs -text
*.bin.* filter=lfs diff=lfs merge=lfs -text
*.bz2 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
*.model filter=lfs diff=lfs merge=lfs -text
*.msgpack 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
*.pt filter=lfs diff=lfs merge=lfs -text
*.pth filter=lfs diff=lfs merge=lfs -text
*.rar filter=lfs diff=lfs merge=lfs -text
saved_model/**/* 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
*.xz filter=lfs diff=lfs merge=lfs -text
*.zip filter=lfs diff=lfs merge=lfs -text
*.zstandard filter=lfs diff=lfs merge=lfs -text
*.tfevents* filter=lfs diff=lfs merge=lfs -text
*.db* filter=lfs diff=lfs merge=lfs -text
*.ark* filter=lfs diff=lfs merge=lfs -text
**/*ckpt*data* filter=lfs diff=lfs merge=lfs -text
**/*ckpt*.meta filter=lfs diff=lfs merge=lfs -text
**/*ckpt*.index filter=lfs diff=lfs merge=lfs -text
*.safetensors filter=lfs diff=lfs merge=lfs -text
*.ckpt filter=lfs diff=lfs merge=lfs -text

44
README.md Normal file
View File

@@ -0,0 +1,44 @@
---
frameworks:
- Pytorch
license: Apache License 2.0
tasks:
- text-generation
---
# 天机-送祝福-7b
天机是一款免费使用、非商业用途的人工智能系统。您可以利用它进行涉及人情世故的任务，如话中有话翻译、说话的艺术建议等，以提升您的情商和核心竞争能力。我们坚信，只有人情世故才是未来AI的核心竞争力，让我们携手见证通用人工智能的来临。 —— "天机不可泄漏。"
Tianji is a free, non-commercial artificial intelligence system. You can utilize it for tasks involving worldly wisdom, such as "subtext translation" and "art of conversation," to enhance your emotional intelligence and core competitiveness. We firmly believe that worldly wisdom are the future core competency of AI, and let us join hands to witness the advent of general artificial intelligence.
SocialAI（来事儿AI） 是设立于中国的非营利组织，我们完全开源了Tianji（天机）系列工作，当前开源模型为送祝福专用微调模型，你可以用它来完成各类送祝福任务。
> *天机地址：https://github.com/SocialAI-tianji*
# 快速开始
```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
mode_name_or_path = '下载好的天机模型地址'
tokenizer = AutoTokenizer.from_pretrained(mode_name_or_path, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(mode_name_or_path, torch_dtype=torch.float16, trust_remote_code=True).cuda().eval()
response, history = model.chat(tokenizer, '祝姐姐考试顺利', meta_instruction='你现在是一个送祝福大师，帮我针对不同人和事情、节日送对应的祝福')
print(response)
```
# 致谢
<div align="center">
***感谢上海人工智能实验室组织的 书生·浦语实战营 学习活动~***
***感谢 OpenXLab 对项目部署的算力支持~***
***感谢 浦语小助手 对项目的支持~***
</div>

36
config.json Normal file
View File

@@ -0,0 +1,36 @@
{
"_name_or_path": "/root/share/model_repos/internlm2-chat-7b",
"architectures": [
"InternLM2ForCausalLM"
],
"attn_implementation": "eager",
"auto_map": {
"AutoConfig": "configuration_internlm.InternLMConfig",
"AutoModel": "modeling_internlm2.InternLM2ForCausalLM",
"AutoModelForCausalLM": "modeling_internlm2.InternLM2ForCausalLM"
},
"bias": false,
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 4096,
"initializer_range": 0.02,
"intermediate_size": 14336,
"max_position_embeddings": 32768,
"model_type": "internlm",
"num_attention_heads": 32,
"num_hidden_layers": 32,
"num_key_value_heads": 8,
"pad_token_id": 2,
"rms_norm_eps": 1e-05,
"rope_scaling": {
"factor": 2.0,
"type": "dynamic"
},
"rope_theta": 1000000,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.37.2",
"use_cache": true,
"vocab_size": 92544
}

1
configuration.json Normal file
View File

@@ -0,0 +1 @@
{"framework":"Pytorch","task":"text-generation"}

164
configuration_internlm.py Normal file
View File

@@ -0,0 +1,164 @@
# coding=utf-8
# Copyright (c) InternLM. 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.configuration_utils import PretrainedConfig
from transformers.utils import logging
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.
num_key_value_heads (`int`, *optional*):
This is the number of key_value heads that should be used to implement Grouped Query Attention. If
`num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
`num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When
converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
by meanpooling all the original heads within that group. For more details checkout [this
paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
`num_attention_heads`.
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__( # pylint: disable=W0102
self,
vocab_size=103168,
hidden_size=4096,
intermediate_size=11008,
num_hidden_layers=32,
num_attention_heads=32,
num_key_value_heads=None,
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,
rope_theta=10000,
rope_scaling=None,
attn_implementation="eager",
**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.bias = bias
if num_key_value_heads is None:
num_key_value_heads = num_attention_heads
self.num_key_value_heads = num_key_value_heads
self.hidden_act = hidden_act
self.initializer_range = initializer_range
self.rms_norm_eps = rms_norm_eps
self.use_cache = use_cache
self.rope_theta = rope_theta
self.rope_scaling = rope_scaling
self._rope_scaling_validation()
self.attn_implementation = attn_implementation
if self.attn_implementation is None:
self.attn_implementation = "eager"
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,
)
def _rope_scaling_validation(self):
"""
Validate the `rope_scaling` configuration.
"""
if self.rope_scaling is None:
return
if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 2:
raise ValueError(
"`rope_scaling` must be a dictionary with with two fields, `type` and `factor`, "
f"got {self.rope_scaling}"
)
rope_scaling_type = self.rope_scaling.get("type", None)
rope_scaling_factor = self.rope_scaling.get("factor", None)
if rope_scaling_type is None or rope_scaling_type not in ["linear", "dynamic"]:
raise ValueError(
f"`rope_scaling`'s type field must be one of ['linear', 'dynamic'], got {rope_scaling_type}"
)
if rope_scaling_factor is None or not isinstance(rope_scaling_factor, float) or rope_scaling_factor < 1.0:
raise ValueError(f"`rope_scaling`'s factor field must be a float >= 1, got {rope_scaling_factor}")

7
generation_config.json Normal file
View File

@@ -0,0 +1,7 @@
{
"_from_model_config": true,
"bos_token_id": 1,
"eos_token_id": 2,
"pad_token_id": 2,
"transformers_version": "4.37.2"
}

3
model-00001-of-00008.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a890611605712ca3c53c003af82c4728a66dfcbcde96fc6386a815a55442cbbf
size 1949337688

3
model-00002-of-00008.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:de6dcfc5365a1e56d21dbaa2ceb89eace6bfea1a27cd54d9771a10b0a4c762ea
size 1946242664

3
model-00003-of-00008.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:761199072d235acf9262777bc451c7c98a9894fec16d1c57734c566d9cb547b0
size 1979780408

3
model-00004-of-00008.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f946bae6256137bbedbf3ece1a3cd75b0a217e2af071c91b86b947dff3e46450
size 1946242696

3
model-00005-of-00008.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:af4ab2ca4694511d79f5eaec2ecfbff0ffa55061e9be1d5bf364e74ad687013c
size 1979780424

3
model-00006-of-00008.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a5696d670352b0b3607061d2f5e5aa8268815e92005cba698ff984353fec979e
size 1946242696

3
model-00007-of-00008.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e4b58acb7817d2f31901e3a789c444121e5505522c9cd1763a2eabed8a70576b
size 1979780424

3
model-00008-of-00008.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f739dda896024086fd850b7e9cac5ab99416de773d0e18d710150425530c2286
size 1748035624

234
model.safetensors.index.json Normal file
View File

@@ -0,0 +1,234 @@
{
"metadata": {
"total_size": 15475417088
},
"weight_map": {
"model.layers.0.attention.wo.weight": "model-00001-of-00008.safetensors",
"model.layers.0.attention.wqkv.weight": "model-00001-of-00008.safetensors",
"model.layers.0.attention_norm.weight": "model-00001-of-00008.safetensors",
"model.layers.0.feed_forward.w1.weight": "model-00001-of-00008.safetensors",
"model.layers.0.feed_forward.w2.weight": "model-00001-of-00008.safetensors",
"model.layers.0.feed_forward.w3.weight": "model-00001-of-00008.safetensors",
"model.layers.0.ffn_norm.weight": "model-00001-of-00008.safetensors",
"model.layers.1.attention.wo.weight": "model-00001-of-00008.safetensors",
"model.layers.1.attention.wqkv.weight": "model-00001-of-00008.safetensors",
"model.layers.1.attention_norm.weight": "model-00001-of-00008.safetensors",
"model.layers.1.feed_forward.w1.weight": "model-00001-of-00008.safetensors",
"model.layers.1.feed_forward.w2.weight": "model-00001-of-00008.safetensors",
"model.layers.1.feed_forward.w3.weight": "model-00001-of-00008.safetensors",
"model.layers.1.ffn_norm.weight": "model-00001-of-00008.safetensors",
"model.layers.10.attention.wo.weight": "model-00003-of-00008.safetensors",
"model.layers.10.attention.wqkv.weight": "model-00003-of-00008.safetensors",
"model.layers.10.attention_norm.weight": "model-00003-of-00008.safetensors",
"model.layers.10.feed_forward.w1.weight": "model-00003-of-00008.safetensors",
"model.layers.10.feed_forward.w2.weight": "model-00003-of-00008.safetensors",
"model.layers.10.feed_forward.w3.weight": "model-00003-of-00008.safetensors",
"model.layers.10.ffn_norm.weight": "model-00003-of-00008.safetensors",
"model.layers.11.attention.wo.weight": "model-00003-of-00008.safetensors",
"model.layers.11.attention.wqkv.weight": "model-00003-of-00008.safetensors",
"model.layers.11.attention_norm.weight": "model-00004-of-00008.safetensors",
"model.layers.11.feed_forward.w1.weight": "model-00003-of-00008.safetensors",
"model.layers.11.feed_forward.w2.weight": "model-00004-of-00008.safetensors",
"model.layers.11.feed_forward.w3.weight": "model-00003-of-00008.safetensors",
"model.layers.11.ffn_norm.weight": "model-00004-of-00008.safetensors",
"model.layers.12.attention.wo.weight": "model-00004-of-00008.safetensors",
"model.layers.12.attention.wqkv.weight": "model-00004-of-00008.safetensors",
"model.layers.12.attention_norm.weight": "model-00004-of-00008.safetensors",
"model.layers.12.feed_forward.w1.weight": "model-00004-of-00008.safetensors",
"model.layers.12.feed_forward.w2.weight": "model-00004-of-00008.safetensors",
"model.layers.12.feed_forward.w3.weight": "model-00004-of-00008.safetensors",
"model.layers.12.ffn_norm.weight": "model-00004-of-00008.safetensors",
"model.layers.13.attention.wo.weight": "model-00004-of-00008.safetensors",
"model.layers.13.attention.wqkv.weight": "model-00004-of-00008.safetensors",
"model.layers.13.attention_norm.weight": "model-00004-of-00008.safetensors",
"model.layers.13.feed_forward.w1.weight": "model-00004-of-00008.safetensors",
"model.layers.13.feed_forward.w2.weight": "model-00004-of-00008.safetensors",
"model.layers.13.feed_forward.w3.weight": "model-00004-of-00008.safetensors",
"model.layers.13.ffn_norm.weight": "model-00004-of-00008.safetensors",
"model.layers.14.attention.wo.weight": "model-00004-of-00008.safetensors",
"model.layers.14.attention.wqkv.weight": "model-00004-of-00008.safetensors",
"model.layers.14.attention_norm.weight": "model-00004-of-00008.safetensors",
"model.layers.14.feed_forward.w1.weight": "model-00004-of-00008.safetensors",
"model.layers.14.feed_forward.w2.weight": "model-00004-of-00008.safetensors",
"model.layers.14.feed_forward.w3.weight": "model-00004-of-00008.safetensors",
"model.layers.14.ffn_norm.weight": "model-00004-of-00008.safetensors",
"model.layers.15.attention.wo.weight": "model-00004-of-00008.safetensors",
"model.layers.15.attention.wqkv.weight": "model-00004-of-00008.safetensors",
"model.layers.15.attention_norm.weight": "model-00004-of-00008.safetensors",
"model.layers.15.feed_forward.w1.weight": "model-00004-of-00008.safetensors",
"model.layers.15.feed_forward.w2.weight": "model-00004-of-00008.safetensors",
"model.layers.15.feed_forward.w3.weight": "model-00004-of-00008.safetensors",
"model.layers.15.ffn_norm.weight": "model-00004-of-00008.safetensors",
"model.layers.16.attention.wo.weight": "model-00004-of-00008.safetensors",
"model.layers.16.attention.wqkv.weight": "model-00004-of-00008.safetensors",
"model.layers.16.attention_norm.weight": "model-00005-of-00008.safetensors",
"model.layers.16.feed_forward.w1.weight": "model-00005-of-00008.safetensors",
"model.layers.16.feed_forward.w2.weight": "model-00005-of-00008.safetensors",
"model.layers.16.feed_forward.w3.weight": "model-00005-of-00008.safetensors",
"model.layers.16.ffn_norm.weight": "model-00005-of-00008.safetensors",
"model.layers.17.attention.wo.weight": "model-00005-of-00008.safetensors",
"model.layers.17.attention.wqkv.weight": "model-00005-of-00008.safetensors",
"model.layers.17.attention_norm.weight": "model-00005-of-00008.safetensors",
"model.layers.17.feed_forward.w1.weight": "model-00005-of-00008.safetensors",
"model.layers.17.feed_forward.w2.weight": "model-00005-of-00008.safetensors",
"model.layers.17.feed_forward.w3.weight": "model-00005-of-00008.safetensors",
"model.layers.17.ffn_norm.weight": "model-00005-of-00008.safetensors",
"model.layers.18.attention.wo.weight": "model-00005-of-00008.safetensors",
"model.layers.18.attention.wqkv.weight": "model-00005-of-00008.safetensors",
"model.layers.18.attention_norm.weight": "model-00005-of-00008.safetensors",
"model.layers.18.feed_forward.w1.weight": "model-00005-of-00008.safetensors",
"model.layers.18.feed_forward.w2.weight": "model-00005-of-00008.safetensors",
"model.layers.18.feed_forward.w3.weight": "model-00005-of-00008.safetensors",
"model.layers.18.ffn_norm.weight": "model-00005-of-00008.safetensors",
"model.layers.19.attention.wo.weight": "model-00005-of-00008.safetensors",
"model.layers.19.attention.wqkv.weight": "model-00005-of-00008.safetensors",
"model.layers.19.attention_norm.weight": "model-00005-of-00008.safetensors",
"model.layers.19.feed_forward.w1.weight": "model-00005-of-00008.safetensors",
"model.layers.19.feed_forward.w2.weight": "model-00005-of-00008.safetensors",
"model.layers.19.feed_forward.w3.weight": "model-00005-of-00008.safetensors",
"model.layers.19.ffn_norm.weight": "model-00005-of-00008.safetensors",
"model.layers.2.attention.wo.weight": "model-00001-of-00008.safetensors",
"model.layers.2.attention.wqkv.weight": "model-00001-of-00008.safetensors",
"model.layers.2.attention_norm.weight": "model-00002-of-00008.safetensors",
"model.layers.2.feed_forward.w1.weight": "model-00001-of-00008.safetensors",
"model.layers.2.feed_forward.w2.weight": "model-00002-of-00008.safetensors",
"model.layers.2.feed_forward.w3.weight": "model-00001-of-00008.safetensors",
"model.layers.2.ffn_norm.weight": "model-00002-of-00008.safetensors",
"model.layers.20.attention.wo.weight": "model-00005-of-00008.safetensors",
"model.layers.20.attention.wqkv.weight": "model-00005-of-00008.safetensors",
"model.layers.20.attention_norm.weight": "model-00006-of-00008.safetensors",
"model.layers.20.feed_forward.w1.weight": "model-00005-of-00008.safetensors",
"model.layers.20.feed_forward.w2.weight": "model-00006-of-00008.safetensors",
"model.layers.20.feed_forward.w3.weight": "model-00005-of-00008.safetensors",
"model.layers.20.ffn_norm.weight": "model-00006-of-00008.safetensors",
"model.layers.21.attention.wo.weight": "model-00006-of-00008.safetensors",
"model.layers.21.attention.wqkv.weight": "model-00006-of-00008.safetensors",
"model.layers.21.attention_norm.weight": "model-00006-of-00008.safetensors",
"model.layers.21.feed_forward.w1.weight": "model-00006-of-00008.safetensors",
"model.layers.21.feed_forward.w2.weight": "model-00006-of-00008.safetensors",
"model.layers.21.feed_forward.w3.weight": "model-00006-of-00008.safetensors",
"model.layers.21.ffn_norm.weight": "model-00006-of-00008.safetensors",
"model.layers.22.attention.wo.weight": "model-00006-of-00008.safetensors",
"model.layers.22.attention.wqkv.weight": "model-00006-of-00008.safetensors",
"model.layers.22.attention_norm.weight": "model-00006-of-00008.safetensors",
"model.layers.22.feed_forward.w1.weight": "model-00006-of-00008.safetensors",
"model.layers.22.feed_forward.w2.weight": "model-00006-of-00008.safetensors",
"model.layers.22.feed_forward.w3.weight": "model-00006-of-00008.safetensors",
"model.layers.22.ffn_norm.weight": "model-00006-of-00008.safetensors",
"model.layers.23.attention.wo.weight": "model-00006-of-00008.safetensors",
"model.layers.23.attention.wqkv.weight": "model-00006-of-00008.safetensors",
"model.layers.23.attention_norm.weight": "model-00006-of-00008.safetensors",
"model.layers.23.feed_forward.w1.weight": "model-00006-of-00008.safetensors",
"model.layers.23.feed_forward.w2.weight": "model-00006-of-00008.safetensors",
"model.layers.23.feed_forward.w3.weight": "model-00006-of-00008.safetensors",
"model.layers.23.ffn_norm.weight": "model-00006-of-00008.safetensors",
"model.layers.24.attention.wo.weight": "model-00006-of-00008.safetensors",
"model.layers.24.attention.wqkv.weight": "model-00006-of-00008.safetensors",
"model.layers.24.attention_norm.weight": "model-00006-of-00008.safetensors",
"model.layers.24.feed_forward.w1.weight": "model-00006-of-00008.safetensors",
"model.layers.24.feed_forward.w2.weight": "model-00006-of-00008.safetensors",
"model.layers.24.feed_forward.w3.weight": "model-00006-of-00008.safetensors",
"model.layers.24.ffn_norm.weight": "model-00006-of-00008.safetensors",
"model.layers.25.attention.wo.weight": "model-00006-of-00008.safetensors",
"model.layers.25.attention.wqkv.weight": "model-00006-of-00008.safetensors",
"model.layers.25.attention_norm.weight": "model-00007-of-00008.safetensors",
"model.layers.25.feed_forward.w1.weight": "model-00007-of-00008.safetensors",
"model.layers.25.feed_forward.w2.weight": "model-00007-of-00008.safetensors",
"model.layers.25.feed_forward.w3.weight": "model-00007-of-00008.safetensors",
"model.layers.25.ffn_norm.weight": "model-00007-of-00008.safetensors",
"model.layers.26.attention.wo.weight": "model-00007-of-00008.safetensors",
"model.layers.26.attention.wqkv.weight": "model-00007-of-00008.safetensors",
"model.layers.26.attention_norm.weight": "model-00007-of-00008.safetensors",
"model.layers.26.feed_forward.w1.weight": "model-00007-of-00008.safetensors",
"model.layers.26.feed_forward.w2.weight": "model-00007-of-00008.safetensors",
"model.layers.26.feed_forward.w3.weight": "model-00007-of-00008.safetensors",
"model.layers.26.ffn_norm.weight": "model-00007-of-00008.safetensors",
"model.layers.27.attention.wo.weight": "model-00007-of-00008.safetensors",
"model.layers.27.attention.wqkv.weight": "model-00007-of-00008.safetensors",
"model.layers.27.attention_norm.weight": "model-00007-of-00008.safetensors",
"model.layers.27.feed_forward.w1.weight": "model-00007-of-00008.safetensors",
"model.layers.27.feed_forward.w2.weight": "model-00007-of-00008.safetensors",
"model.layers.27.feed_forward.w3.weight": "model-00007-of-00008.safetensors",
"model.layers.27.ffn_norm.weight": "model-00007-of-00008.safetensors",
"model.layers.28.attention.wo.weight": "model-00007-of-00008.safetensors",
"model.layers.28.attention.wqkv.weight": "model-00007-of-00008.safetensors",
"model.layers.28.attention_norm.weight": "model-00007-of-00008.safetensors",
"model.layers.28.feed_forward.w1.weight": "model-00007-of-00008.safetensors",
"model.layers.28.feed_forward.w2.weight": "model-00007-of-00008.safetensors",
"model.layers.28.feed_forward.w3.weight": "model-00007-of-00008.safetensors",
"model.layers.28.ffn_norm.weight": "model-00007-of-00008.safetensors",
"model.layers.29.attention.wo.weight": "model-00007-of-00008.safetensors",
"model.layers.29.attention.wqkv.weight": "model-00007-of-00008.safetensors",
"model.layers.29.attention_norm.weight": "model-00008-of-00008.safetensors",
"model.layers.29.feed_forward.w1.weight": "model-00007-of-00008.safetensors",
"model.layers.29.feed_forward.w2.weight": "model-00008-of-00008.safetensors",
"model.layers.29.feed_forward.w3.weight": "model-00007-of-00008.safetensors",
"model.layers.29.ffn_norm.weight": "model-00008-of-00008.safetensors",
"model.layers.3.attention.wo.weight": "model-00002-of-00008.safetensors",
"model.layers.3.attention.wqkv.weight": "model-00002-of-00008.safetensors",
"model.layers.3.attention_norm.weight": "model-00002-of-00008.safetensors",
"model.layers.3.feed_forward.w1.weight": "model-00002-of-00008.safetensors",
"model.layers.3.feed_forward.w2.weight": "model-00002-of-00008.safetensors",
"model.layers.3.feed_forward.w3.weight": "model-00002-of-00008.safetensors",
"model.layers.3.ffn_norm.weight": "model-00002-of-00008.safetensors",
"model.layers.30.attention.wo.weight": "model-00008-of-00008.safetensors",
"model.layers.30.attention.wqkv.weight": "model-00008-of-00008.safetensors",
"model.layers.30.attention_norm.weight": "model-00008-of-00008.safetensors",
"model.layers.30.feed_forward.w1.weight": "model-00008-of-00008.safetensors",
"model.layers.30.feed_forward.w2.weight": "model-00008-of-00008.safetensors",
"model.layers.30.feed_forward.w3.weight": "model-00008-of-00008.safetensors",
"model.layers.30.ffn_norm.weight": "model-00008-of-00008.safetensors",
"model.layers.31.attention.wo.weight": "model-00008-of-00008.safetensors",
"model.layers.31.attention.wqkv.weight": "model-00008-of-00008.safetensors",
"model.layers.31.attention_norm.weight": "model-00008-of-00008.safetensors",
"model.layers.31.feed_forward.w1.weight": "model-00008-of-00008.safetensors",
"model.layers.31.feed_forward.w2.weight": "model-00008-of-00008.safetensors",
"model.layers.31.feed_forward.w3.weight": "model-00008-of-00008.safetensors",
"model.layers.31.ffn_norm.weight": "model-00008-of-00008.safetensors",
"model.layers.4.attention.wo.weight": "model-00002-of-00008.safetensors",
"model.layers.4.attention.wqkv.weight": "model-00002-of-00008.safetensors",
"model.layers.4.attention_norm.weight": "model-00002-of-00008.safetensors",
"model.layers.4.feed_forward.w1.weight": "model-00002-of-00008.safetensors",
"model.layers.4.feed_forward.w2.weight": "model-00002-of-00008.safetensors",
"model.layers.4.feed_forward.w3.weight": "model-00002-of-00008.safetensors",
"model.layers.4.ffn_norm.weight": "model-00002-of-00008.safetensors",
"model.layers.5.attention.wo.weight": "model-00002-of-00008.safetensors",
"model.layers.5.attention.wqkv.weight": "model-00002-of-00008.safetensors",
"model.layers.5.attention_norm.weight": "model-00002-of-00008.safetensors",
"model.layers.5.feed_forward.w1.weight": "model-00002-of-00008.safetensors",
"model.layers.5.feed_forward.w2.weight": "model-00002-of-00008.safetensors",
"model.layers.5.feed_forward.w3.weight": "model-00002-of-00008.safetensors",
"model.layers.5.ffn_norm.weight": "model-00002-of-00008.safetensors",
"model.layers.6.attention.wo.weight": "model-00002-of-00008.safetensors",
"model.layers.6.attention.wqkv.weight": "model-00002-of-00008.safetensors",
"model.layers.6.attention_norm.weight": "model-00002-of-00008.safetensors",
"model.layers.6.feed_forward.w1.weight": "model-00002-of-00008.safetensors",
"model.layers.6.feed_forward.w2.weight": "model-00002-of-00008.safetensors",
"model.layers.6.feed_forward.w3.weight": "model-00002-of-00008.safetensors",
"model.layers.6.ffn_norm.weight": "model-00002-of-00008.safetensors",
"model.layers.7.attention.wo.weight": "model-00002-of-00008.safetensors",
"model.layers.7.attention.wqkv.weight": "model-00002-of-00008.safetensors",
"model.layers.7.attention_norm.weight": "model-00003-of-00008.safetensors",
"model.layers.7.feed_forward.w1.weight": "model-00003-of-00008.safetensors",
"model.layers.7.feed_forward.w2.weight": "model-00003-of-00008.safetensors",
"model.layers.7.feed_forward.w3.weight": "model-00003-of-00008.safetensors",
"model.layers.7.ffn_norm.weight": "model-00003-of-00008.safetensors",
"model.layers.8.attention.wo.weight": "model-00003-of-00008.safetensors",
"model.layers.8.attention.wqkv.weight": "model-00003-of-00008.safetensors",
"model.layers.8.attention_norm.weight": "model-00003-of-00008.safetensors",
"model.layers.8.feed_forward.w1.weight": "model-00003-of-00008.safetensors",
"model.layers.8.feed_forward.w2.weight": "model-00003-of-00008.safetensors",
"model.layers.8.feed_forward.w3.weight": "model-00003-of-00008.safetensors",
"model.layers.8.ffn_norm.weight": "model-00003-of-00008.safetensors",
"model.layers.9.attention.wo.weight": "model-00003-of-00008.safetensors",
"model.layers.9.attention.wqkv.weight": "model-00003-of-00008.safetensors",
"model.layers.9.attention_norm.weight": "model-00003-of-00008.safetensors",
"model.layers.9.feed_forward.w1.weight": "model-00003-of-00008.safetensors",
"model.layers.9.feed_forward.w2.weight": "model-00003-of-00008.safetensors",
"model.layers.9.feed_forward.w3.weight": "model-00003-of-00008.safetensors",
"model.layers.9.ffn_norm.weight": "model-00003-of-00008.safetensors",
"model.norm.weight": "model-00008-of-00008.safetensors",
"model.tok_embeddings.weight": "model-00001-of-00008.safetensors",
"output.weight": "model-00008-of-00008.safetensors"
}
}

1385
modeling_internlm2.py Normal file
View File

File diff suppressed because it is too large Load Diff

6
special_tokens_map.json Normal file
View File

@@ -0,0 +1,6 @@
{
"bos_token": "<s>",
"eos_token": "</s>",
"pad_token": "</s>",
"unk_token": "<unk>"
}

240
tokenization_internlm.py Normal file
View File

@@ -0,0 +1,240 @@
# coding=utf-8
# Copyright (c) InternLM. 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,
)
""" Initialization"""
@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
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f868398fc4e05ee1e8aeba95ddf18ddcc45b8bce55d5093bead5bbf80429b48b
size 1477754

90
tokenizer_config.json Normal file
View File

@@ -0,0 +1,90 @@
{
"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",
"unk_token": "<unk>",
"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
},
"92543": {
"content": "<|im_start|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"92542": {
"content": "<|im_end|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"92541": {
"content": "<|action_start|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"92540": {
"content": "<|action_end|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"92539": {
"content": "<|interpreter|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"92538": {
"content": "<|plugin|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
}
},
"chat_template": "{{ bos_token }}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}"
}