Duplicate from togethercomputer/Pythia-Chat-Base-7B
Browse filesCo-authored-by: Jue Wang <[email protected]>
- .gitattributes +34 -0
- README.md +177 -0
- config.json +25 -0
- pytorch_model-00001-of-00002.bin +3 -0
- pytorch_model-00002-of-00002.bin +3 -0
- pytorch_model.bin.index.json +491 -0
- special_tokens_map.json +5 -0
- tokenizer.json +0 -0
- tokenizer_config.json +9 -0
.gitattributes
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
28 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
29 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
30 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
31 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
32 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
33 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
34 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,177 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
language:
|
4 |
+
- en
|
5 |
+
duplicated_from: togethercomputer/Pythia-Chat-Base-7B
|
6 |
+
---
|
7 |
+
|
8 |
+
***<p style="font-size: 24px">Feel free to try out our [OpenChatKit feedback app](https://huggingface.co/spaces/togethercomputer/OpenChatKit)!</p>***
|
9 |
+
|
10 |
+
# Pythia-Chat-Base-7B-v0.16
|
11 |
+
|
12 |
+
> TLDR: As part of OpenChatKit (codebase available [here](https://github.com/togethercomputer/OpenChaT)),
|
13 |
+
> Pythia-Chat-Base-7B-v0.16 is a 7B parameter language model, fine-tuned from EleutherAI’s Pythia 7B with over 40 million instructions on 100% carbon negative compute.
|
14 |
+
|
15 |
+
Pythia-Chat-Base-7B-v0.16 is based on ElutherAI’s Pythia-7B model, and is fine-tuned with data focusing on dialog-style interactions.
|
16 |
+
We focused the tuning on several tasks such as question answering, classification, extraction, and summarization.
|
17 |
+
We’ve fine-tuned the model with a collection of 43 million high-quality instructions.
|
18 |
+
Together partnered with LAION and Ontocord.ai, who both helped curate the dataset the model is based on.
|
19 |
+
You can read more about this process and the availability of this dataset in LAION’s blog post [here](https://laion.ai/blog/oig-dataset/).
|
20 |
+
|
21 |
+
In addition to the aforementioned fine-tuning, Pythia-Chat-Base-7B-v0.16 has also undergone further fine-tuning via a small amount of feedback data.
|
22 |
+
This process allows the model to better adapt to human preferences in the conversations.
|
23 |
+
|
24 |
+
One of the notable features of Pythia-Chat-Base-7B-v0.16 is its ability to **run inference on a 12GB GPU**, thanks to the quantization technique.
|
25 |
+
It helps maintain the dialogue capabilities while making the model more accessible to a wider range of users and hardware configurations.
|
26 |
+
|
27 |
+
## Model Details
|
28 |
+
- **Developed by**: Together Computer.
|
29 |
+
- **Model type**: Language Model
|
30 |
+
- **Language(s)**: English
|
31 |
+
- **License**: Apache 2.0
|
32 |
+
- **Model Description**: A 7B parameter open source chat model, fine-tuned from EleutherAI’s Pythia with over 40M instructions on 100% carbon negative compute
|
33 |
+
- **Resources for more information**: [GitHub Repository](https://github.com/togethercomputer/OpenChaT).
|
34 |
+
|
35 |
+
# Quick Start
|
36 |
+
|
37 |
+
## GPU Inference
|
38 |
+
|
39 |
+
This requires a GPU with 24GB memory.
|
40 |
+
```python
|
41 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
42 |
+
|
43 |
+
# init
|
44 |
+
tokenizer = AutoTokenizer.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16")
|
45 |
+
model = AutoModelForCausalLM.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16", torch_dtype=torch.float16)
|
46 |
+
model = model.to('cuda:0')
|
47 |
+
|
48 |
+
# infer
|
49 |
+
inputs = tokenizer("<human>: Hello!\n<bot>:", return_tensors='pt').to(model.device)
|
50 |
+
outputs = model.generate(**inputs, max_new_tokens=10, do_sample=True, temperature=0.8)
|
51 |
+
output_str = tokenizer.decode(outputs[0])
|
52 |
+
print(output_str)
|
53 |
+
```
|
54 |
+
|
55 |
+
## GPU Inference in Int8
|
56 |
+
|
57 |
+
This requires a GPU with 12GB memory.
|
58 |
+
```python
|
59 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
60 |
+
|
61 |
+
# init
|
62 |
+
tokenizer = AutoTokenizer.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16")
|
63 |
+
model = AutoModelForCausalLM.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16", device_map="auto", load_in_8bit=True)
|
64 |
+
|
65 |
+
# infer
|
66 |
+
inputs = tokenizer("<human>: Hello!\n<bot>:", return_tensors='pt').to(model.device)
|
67 |
+
outputs = model.generate(**inputs, max_new_tokens=10, do_sample=True, temperature=0.8)
|
68 |
+
output_str = tokenizer.decode(outputs[0])
|
69 |
+
print(output_str)
|
70 |
+
```
|
71 |
+
|
72 |
+
|
73 |
+
## CPU Inference
|
74 |
+
|
75 |
+
```python
|
76 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
77 |
+
|
78 |
+
# init
|
79 |
+
tokenizer = AutoTokenizer.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16")
|
80 |
+
model = AutoModelForCausalLM.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16", torch_dtype=torch.bfloat16)
|
81 |
+
|
82 |
+
# infer
|
83 |
+
inputs = tokenizer("<human>: Hello!\n<bot>:", return_tensors='pt').to(model.device)
|
84 |
+
outputs = model.generate(**inputs, max_new_tokens=10, do_sample=True, temperature=0.8)
|
85 |
+
output_str = tokenizer.decode(outputs[0])
|
86 |
+
print(output_str)
|
87 |
+
```
|
88 |
+
|
89 |
+
|
90 |
+
## Strengths of the model
|
91 |
+
|
92 |
+
There are several tasks that OpenChatKit excels at out of the box. This includes:
|
93 |
+
|
94 |
+
- Summarization and question answering within context.
|
95 |
+
- Extraction.
|
96 |
+
- Classification.
|
97 |
+
|
98 |
+
In addition, the model does well on few-shot prompts. For both classification and extraction, the model performs even better with few shots, as in most HELM tasks. [Contact us](https://www.together.xyz/contact) if you’re interested in trying few-shot prompts with the model.
|
99 |
+
|
100 |
+
## Weaknesses of the model
|
101 |
+
|
102 |
+
That said, there are several areas where we have more work to do, and we need your help! Some of these include:
|
103 |
+
|
104 |
+
- Knowledge-based closed question and answering: The chatbot may hallucinate and give incorrect results. Be sure to fact check, and if possible provide feedback with the corrected information.
|
105 |
+
- Coding tasks: The chatbot was not trained on a large enough corpus of source code to excel at writing code. We welcome contributions of additional datasets to improve this!
|
106 |
+
- Repetition: Sometimes the chatbot will repeat its response. We’re working to improve this, but in the meantime you can click the refresh button to start a new conversation.
|
107 |
+
- Context switching: If you change the topic in the middle of a conversation the chatbot often cannot make the switch automatically and will continue to give answers related to the prior topic.
|
108 |
+
- Creative writing and longer answers: The chatbot does not generate long, creative text such as an essay or story.
|
109 |
+
|
110 |
+
We are excited to work with you to address these weaknesses by getting your feedback, bolstering data sets, and improving accuracy.
|
111 |
+
|
112 |
+
# Uses
|
113 |
+
|
114 |
+
## Direct Use
|
115 |
+
|
116 |
+
The model is intended for research purposes. Possible research areas and tasks include
|
117 |
+
|
118 |
+
- Safe deployment of models which have the potential to generate harmful content.
|
119 |
+
- Probing and understanding the limitations and biases of dialogue models or language models.
|
120 |
+
- Generation of artworks and use in design and other artistic processes.
|
121 |
+
- Applications in educational or creative tools.
|
122 |
+
- Research on dialogue models or language models.
|
123 |
+
|
124 |
+
Excluded uses are described below.
|
125 |
+
|
126 |
+
### Misuse, Malicious Use, and Out-of-Scope Use
|
127 |
+
|
128 |
+
The OpenChatKit community provides Pythia-Chat-Base-7B-v0.16 as an open source tool for building chatbots.
|
129 |
+
The community is not responsible for any misuse, malicious use, or out-of-scope use of the model.
|
130 |
+
It is the responsibility of the end user to ensure that the model is used in a responsible and ethical manner.
|
131 |
+
|
132 |
+
#### Out-of-Scope Use
|
133 |
+
|
134 |
+
Pythia-Chat-Base-7B-v0.16 is designed for use in chatbot applications and may not perform well for other use cases outside of its intended scope.
|
135 |
+
For example, it may not be suitable for use in safety-critical applications or for making decisions that have a significant impact on individuals or society.
|
136 |
+
It is important to consider the limitations of the model and to only use it for its intended purpose.
|
137 |
+
|
138 |
+
#### Misuse and Malicious Use
|
139 |
+
|
140 |
+
Pythia-Chat-Base-7B-v0.16 is designed for use in chatbot applications and should not be used for any other purpose.
|
141 |
+
Misuse of the model, such as using it to engage in illegal or unethical activities, is strictly prohibited and goes against the principles of the OpenChatKit community project.
|
142 |
+
|
143 |
+
Using the model to generate content that is cruel to individuals is a misuse of this model. This includes, but is not limited to:
|
144 |
+
|
145 |
+
- Generating fake news, misinformation, or propaganda
|
146 |
+
- Promoting hate speech, discrimination, or violence against individuals or groups
|
147 |
+
- Impersonating individuals or organizations without their consent
|
148 |
+
- Engaging in cyberbullying or harassment
|
149 |
+
- Defamatory content
|
150 |
+
- Spamming or scamming
|
151 |
+
- Sharing confidential or sensitive information without proper authorization
|
152 |
+
- Violating the terms of use of the model or the data used to train it
|
153 |
+
- Creating automated bots for malicious purposes such as spreading malware, phishing scams, or spamming
|
154 |
+
|
155 |
+
## Limitations
|
156 |
+
|
157 |
+
Pythia-Chat-Base-7B-v0.16, like other language model-based chatbots, has limitations that should be taken into consideration.
|
158 |
+
For example, the model may not always provide accurate or relevant answers, particularly for questions that are complex, ambiguous, or outside of its training data.
|
159 |
+
We therefore welcome contributions from individuals and organizations, and encourage collaboration towards creating a more robust and inclusive chatbot.
|
160 |
+
|
161 |
+
## Training
|
162 |
+
|
163 |
+
**Training Data**
|
164 |
+
|
165 |
+
Please refer to [togethercomputer/OpenDataHub](https://github.com/togethercomputer/OpenDataHub)
|
166 |
+
|
167 |
+
**Training Procedure**
|
168 |
+
|
169 |
+
- **Hardware:** 8 x A100 GPUs
|
170 |
+
- **Optimizer:** [8bit-AdamW](https://github.com/TimDettmers/bitsandbytes)
|
171 |
+
- **Gradient Accumulations**: 4
|
172 |
+
- **Batch:** 4 x 4 x 16 x 2048 = 524288 tokens
|
173 |
+
- **Learning rate:** warmup to 1e-5 for 100 steps and then kept constant
|
174 |
+
|
175 |
+
## Community
|
176 |
+
|
177 |
+
Join us on [Together Discord](https://discord.gg/6ZVDU8tTD4)
|
config.json
ADDED
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "togethercomputer/Pythia-Chat-Base-7B-v0.16",
|
3 |
+
"architectures": [
|
4 |
+
"GPTNeoXForCausalLM"
|
5 |
+
],
|
6 |
+
"bos_token_id": 0,
|
7 |
+
"eos_token_id": 0,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_size": 4096,
|
10 |
+
"initializer_range": 0.02,
|
11 |
+
"intermediate_size": 16384,
|
12 |
+
"layer_norm_eps": 1e-05,
|
13 |
+
"max_position_embeddings": 2048,
|
14 |
+
"model_type": "gpt_neox",
|
15 |
+
"num_attention_heads": 32,
|
16 |
+
"num_hidden_layers": 32,
|
17 |
+
"rotary_emb_base": 10000,
|
18 |
+
"rotary_pct": 0.25,
|
19 |
+
"tie_word_embeddings": false,
|
20 |
+
"torch_dtype": "float16",
|
21 |
+
"transformers_version": "4.21.1",
|
22 |
+
"use_cache": true,
|
23 |
+
"use_parallel_residual": true,
|
24 |
+
"vocab_size": 50432
|
25 |
+
}
|
pytorch_model-00001-of-00002.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:76c64e60ad71d8835fbcf6be49321594f5c02b85f0e3d8cda8b6781d1531b321
|
3 |
+
size 10045933972
|
pytorch_model-00002-of-00002.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7ac9e162f32e582db7a54b65df40f73a4e0243f61597e6cf212369030ac13e6f
|
3 |
+
size 3803055346
|
pytorch_model.bin.index.json
ADDED
@@ -0,0 +1,491 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"metadata": {
|
3 |
+
"total_size": 13848822848
|
4 |
+
},
|
5 |
+
"weight_map": {
|
6 |
+
"embed_out.weight": "pytorch_model-00002-of-00002.bin",
|
7 |
+
"gpt_neox.embed_in.weight": "pytorch_model-00001-of-00002.bin",
|
8 |
+
"gpt_neox.final_layer_norm.bias": "pytorch_model-00002-of-00002.bin",
|
9 |
+
"gpt_neox.final_layer_norm.weight": "pytorch_model-00002-of-00002.bin",
|
10 |
+
"gpt_neox.layers.0.attention.bias": "pytorch_model-00001-of-00002.bin",
|
11 |
+
"gpt_neox.layers.0.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
12 |
+
"gpt_neox.layers.0.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
13 |
+
"gpt_neox.layers.0.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
14 |
+
"gpt_neox.layers.0.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
15 |
+
"gpt_neox.layers.0.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
16 |
+
"gpt_neox.layers.0.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
17 |
+
"gpt_neox.layers.0.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
18 |
+
"gpt_neox.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
19 |
+
"gpt_neox.layers.0.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
20 |
+
"gpt_neox.layers.0.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
21 |
+
"gpt_neox.layers.0.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
22 |
+
"gpt_neox.layers.0.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
23 |
+
"gpt_neox.layers.0.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
24 |
+
"gpt_neox.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
25 |
+
"gpt_neox.layers.1.attention.bias": "pytorch_model-00001-of-00002.bin",
|
26 |
+
"gpt_neox.layers.1.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
27 |
+
"gpt_neox.layers.1.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
28 |
+
"gpt_neox.layers.1.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
29 |
+
"gpt_neox.layers.1.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
30 |
+
"gpt_neox.layers.1.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
31 |
+
"gpt_neox.layers.1.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
32 |
+
"gpt_neox.layers.1.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
33 |
+
"gpt_neox.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
34 |
+
"gpt_neox.layers.1.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
35 |
+
"gpt_neox.layers.1.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
36 |
+
"gpt_neox.layers.1.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
37 |
+
"gpt_neox.layers.1.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
38 |
+
"gpt_neox.layers.1.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
39 |
+
"gpt_neox.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
40 |
+
"gpt_neox.layers.10.attention.bias": "pytorch_model-00001-of-00002.bin",
|
41 |
+
"gpt_neox.layers.10.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
42 |
+
"gpt_neox.layers.10.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
43 |
+
"gpt_neox.layers.10.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
44 |
+
"gpt_neox.layers.10.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
45 |
+
"gpt_neox.layers.10.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
46 |
+
"gpt_neox.layers.10.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
47 |
+
"gpt_neox.layers.10.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
48 |
+
"gpt_neox.layers.10.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
49 |
+
"gpt_neox.layers.10.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
50 |
+
"gpt_neox.layers.10.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
51 |
+
"gpt_neox.layers.10.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
52 |
+
"gpt_neox.layers.10.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
53 |
+
"gpt_neox.layers.10.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
54 |
+
"gpt_neox.layers.10.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
55 |
+
"gpt_neox.layers.11.attention.bias": "pytorch_model-00001-of-00002.bin",
|
56 |
+
"gpt_neox.layers.11.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
57 |
+
"gpt_neox.layers.11.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
58 |
+
"gpt_neox.layers.11.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
59 |
+
"gpt_neox.layers.11.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
60 |
+
"gpt_neox.layers.11.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
61 |
+
"gpt_neox.layers.11.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
62 |
+
"gpt_neox.layers.11.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
63 |
+
"gpt_neox.layers.11.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
64 |
+
"gpt_neox.layers.11.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
65 |
+
"gpt_neox.layers.11.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
66 |
+
"gpt_neox.layers.11.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
67 |
+
"gpt_neox.layers.11.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
68 |
+
"gpt_neox.layers.11.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
69 |
+
"gpt_neox.layers.11.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
70 |
+
"gpt_neox.layers.12.attention.bias": "pytorch_model-00001-of-00002.bin",
|
71 |
+
"gpt_neox.layers.12.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
72 |
+
"gpt_neox.layers.12.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
73 |
+
"gpt_neox.layers.12.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
74 |
+
"gpt_neox.layers.12.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
75 |
+
"gpt_neox.layers.12.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
76 |
+
"gpt_neox.layers.12.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
77 |
+
"gpt_neox.layers.12.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
78 |
+
"gpt_neox.layers.12.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
79 |
+
"gpt_neox.layers.12.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
80 |
+
"gpt_neox.layers.12.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
81 |
+
"gpt_neox.layers.12.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
82 |
+
"gpt_neox.layers.12.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
83 |
+
"gpt_neox.layers.12.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
84 |
+
"gpt_neox.layers.12.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
85 |
+
"gpt_neox.layers.13.attention.bias": "pytorch_model-00001-of-00002.bin",
|
86 |
+
"gpt_neox.layers.13.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
87 |
+
"gpt_neox.layers.13.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
88 |
+
"gpt_neox.layers.13.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
89 |
+
"gpt_neox.layers.13.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
90 |
+
"gpt_neox.layers.13.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
91 |
+
"gpt_neox.layers.13.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
92 |
+
"gpt_neox.layers.13.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
93 |
+
"gpt_neox.layers.13.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
94 |
+
"gpt_neox.layers.13.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
95 |
+
"gpt_neox.layers.13.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
96 |
+
"gpt_neox.layers.13.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
97 |
+
"gpt_neox.layers.13.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
98 |
+
"gpt_neox.layers.13.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
99 |
+
"gpt_neox.layers.13.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
100 |
+
"gpt_neox.layers.14.attention.bias": "pytorch_model-00001-of-00002.bin",
|
101 |
+
"gpt_neox.layers.14.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
102 |
+
"gpt_neox.layers.14.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
103 |
+
"gpt_neox.layers.14.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
104 |
+
"gpt_neox.layers.14.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
105 |
+
"gpt_neox.layers.14.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
106 |
+
"gpt_neox.layers.14.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
107 |
+
"gpt_neox.layers.14.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
108 |
+
"gpt_neox.layers.14.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
109 |
+
"gpt_neox.layers.14.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
110 |
+
"gpt_neox.layers.14.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
111 |
+
"gpt_neox.layers.14.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
112 |
+
"gpt_neox.layers.14.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
113 |
+
"gpt_neox.layers.14.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
114 |
+
"gpt_neox.layers.14.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
115 |
+
"gpt_neox.layers.15.attention.bias": "pytorch_model-00001-of-00002.bin",
|
116 |
+
"gpt_neox.layers.15.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
117 |
+
"gpt_neox.layers.15.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
118 |
+
"gpt_neox.layers.15.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
119 |
+
"gpt_neox.layers.15.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
120 |
+
"gpt_neox.layers.15.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
121 |
+
"gpt_neox.layers.15.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
122 |
+
"gpt_neox.layers.15.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
123 |
+
"gpt_neox.layers.15.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
124 |
+
"gpt_neox.layers.15.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
125 |
+
"gpt_neox.layers.15.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
126 |
+
"gpt_neox.layers.15.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
127 |
+
"gpt_neox.layers.15.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
128 |
+
"gpt_neox.layers.15.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
129 |
+
"gpt_neox.layers.15.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
130 |
+
"gpt_neox.layers.16.attention.bias": "pytorch_model-00001-of-00002.bin",
|
131 |
+
"gpt_neox.layers.16.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
132 |
+
"gpt_neox.layers.16.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
133 |
+
"gpt_neox.layers.16.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
134 |
+
"gpt_neox.layers.16.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
135 |
+
"gpt_neox.layers.16.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
136 |
+
"gpt_neox.layers.16.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
137 |
+
"gpt_neox.layers.16.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
138 |
+
"gpt_neox.layers.16.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
139 |
+
"gpt_neox.layers.16.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
140 |
+
"gpt_neox.layers.16.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
141 |
+
"gpt_neox.layers.16.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
142 |
+
"gpt_neox.layers.16.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
143 |
+
"gpt_neox.layers.16.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
144 |
+
"gpt_neox.layers.16.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
145 |
+
"gpt_neox.layers.17.attention.bias": "pytorch_model-00001-of-00002.bin",
|
146 |
+
"gpt_neox.layers.17.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
147 |
+
"gpt_neox.layers.17.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
148 |
+
"gpt_neox.layers.17.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
149 |
+
"gpt_neox.layers.17.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
150 |
+
"gpt_neox.layers.17.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
151 |
+
"gpt_neox.layers.17.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
152 |
+
"gpt_neox.layers.17.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
153 |
+
"gpt_neox.layers.17.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
154 |
+
"gpt_neox.layers.17.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
155 |
+
"gpt_neox.layers.17.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
156 |
+
"gpt_neox.layers.17.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
157 |
+
"gpt_neox.layers.17.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
158 |
+
"gpt_neox.layers.17.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
159 |
+
"gpt_neox.layers.17.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
160 |
+
"gpt_neox.layers.18.attention.bias": "pytorch_model-00001-of-00002.bin",
|
161 |
+
"gpt_neox.layers.18.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
162 |
+
"gpt_neox.layers.18.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
163 |
+
"gpt_neox.layers.18.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
164 |
+
"gpt_neox.layers.18.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
165 |
+
"gpt_neox.layers.18.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
166 |
+
"gpt_neox.layers.18.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
167 |
+
"gpt_neox.layers.18.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
168 |
+
"gpt_neox.layers.18.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
169 |
+
"gpt_neox.layers.18.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
170 |
+
"gpt_neox.layers.18.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
171 |
+
"gpt_neox.layers.18.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
172 |
+
"gpt_neox.layers.18.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
173 |
+
"gpt_neox.layers.18.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
174 |
+
"gpt_neox.layers.18.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
175 |
+
"gpt_neox.layers.19.attention.bias": "pytorch_model-00001-of-00002.bin",
|
176 |
+
"gpt_neox.layers.19.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
177 |
+
"gpt_neox.layers.19.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
178 |
+
"gpt_neox.layers.19.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
179 |
+
"gpt_neox.layers.19.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
180 |
+
"gpt_neox.layers.19.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
181 |
+
"gpt_neox.layers.19.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
182 |
+
"gpt_neox.layers.19.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
183 |
+
"gpt_neox.layers.19.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
184 |
+
"gpt_neox.layers.19.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
185 |
+
"gpt_neox.layers.19.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
186 |
+
"gpt_neox.layers.19.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
187 |
+
"gpt_neox.layers.19.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
188 |
+
"gpt_neox.layers.19.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
189 |
+
"gpt_neox.layers.19.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
190 |
+
"gpt_neox.layers.2.attention.bias": "pytorch_model-00001-of-00002.bin",
|
191 |
+
"gpt_neox.layers.2.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
192 |
+
"gpt_neox.layers.2.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
193 |
+
"gpt_neox.layers.2.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
194 |
+
"gpt_neox.layers.2.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
195 |
+
"gpt_neox.layers.2.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
196 |
+
"gpt_neox.layers.2.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
197 |
+
"gpt_neox.layers.2.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
198 |
+
"gpt_neox.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
199 |
+
"gpt_neox.layers.2.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
200 |
+
"gpt_neox.layers.2.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
201 |
+
"gpt_neox.layers.2.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
202 |
+
"gpt_neox.layers.2.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
203 |
+
"gpt_neox.layers.2.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
204 |
+
"gpt_neox.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
205 |
+
"gpt_neox.layers.20.attention.bias": "pytorch_model-00001-of-00002.bin",
|
206 |
+
"gpt_neox.layers.20.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
207 |
+
"gpt_neox.layers.20.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
208 |
+
"gpt_neox.layers.20.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
209 |
+
"gpt_neox.layers.20.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
210 |
+
"gpt_neox.layers.20.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
211 |
+
"gpt_neox.layers.20.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
212 |
+
"gpt_neox.layers.20.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
213 |
+
"gpt_neox.layers.20.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
214 |
+
"gpt_neox.layers.20.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
215 |
+
"gpt_neox.layers.20.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
216 |
+
"gpt_neox.layers.20.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
217 |
+
"gpt_neox.layers.20.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
218 |
+
"gpt_neox.layers.20.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
219 |
+
"gpt_neox.layers.20.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
220 |
+
"gpt_neox.layers.21.attention.bias": "pytorch_model-00001-of-00002.bin",
|
221 |
+
"gpt_neox.layers.21.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
222 |
+
"gpt_neox.layers.21.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
223 |
+
"gpt_neox.layers.21.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
224 |
+
"gpt_neox.layers.21.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
225 |
+
"gpt_neox.layers.21.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
226 |
+
"gpt_neox.layers.21.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
227 |
+
"gpt_neox.layers.21.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
228 |
+
"gpt_neox.layers.21.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
229 |
+
"gpt_neox.layers.21.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
230 |
+
"gpt_neox.layers.21.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
231 |
+
"gpt_neox.layers.21.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
232 |
+
"gpt_neox.layers.21.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
233 |
+
"gpt_neox.layers.21.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
234 |
+
"gpt_neox.layers.21.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
235 |
+
"gpt_neox.layers.22.attention.bias": "pytorch_model-00001-of-00002.bin",
|
236 |
+
"gpt_neox.layers.22.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
237 |
+
"gpt_neox.layers.22.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
238 |
+
"gpt_neox.layers.22.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
239 |
+
"gpt_neox.layers.22.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
240 |
+
"gpt_neox.layers.22.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
241 |
+
"gpt_neox.layers.22.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
242 |
+
"gpt_neox.layers.22.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
243 |
+
"gpt_neox.layers.22.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
244 |
+
"gpt_neox.layers.22.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
245 |
+
"gpt_neox.layers.22.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
246 |
+
"gpt_neox.layers.22.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
247 |
+
"gpt_neox.layers.22.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
248 |
+
"gpt_neox.layers.22.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
249 |
+
"gpt_neox.layers.22.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
250 |
+
"gpt_neox.layers.23.attention.bias": "pytorch_model-00001-of-00002.bin",
|
251 |
+
"gpt_neox.layers.23.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
252 |
+
"gpt_neox.layers.23.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
253 |
+
"gpt_neox.layers.23.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
254 |
+
"gpt_neox.layers.23.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
255 |
+
"gpt_neox.layers.23.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
256 |
+
"gpt_neox.layers.23.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
257 |
+
"gpt_neox.layers.23.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
258 |
+
"gpt_neox.layers.23.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
259 |
+
"gpt_neox.layers.23.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
|
260 |
+
"gpt_neox.layers.23.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
|
261 |
+
"gpt_neox.layers.23.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
|
262 |
+
"gpt_neox.layers.23.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
|
263 |
+
"gpt_neox.layers.23.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
264 |
+
"gpt_neox.layers.23.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
265 |
+
"gpt_neox.layers.24.attention.bias": "pytorch_model-00002-of-00002.bin",
|
266 |
+
"gpt_neox.layers.24.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
|
267 |
+
"gpt_neox.layers.24.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
|
268 |
+
"gpt_neox.layers.24.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
|
269 |
+
"gpt_neox.layers.24.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
|
270 |
+
"gpt_neox.layers.24.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
|
271 |
+
"gpt_neox.layers.24.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
|
272 |
+
"gpt_neox.layers.24.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
273 |
+
"gpt_neox.layers.24.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
274 |
+
"gpt_neox.layers.24.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
|
275 |
+
"gpt_neox.layers.24.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
|
276 |
+
"gpt_neox.layers.24.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
|
277 |
+
"gpt_neox.layers.24.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
|
278 |
+
"gpt_neox.layers.24.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
279 |
+
"gpt_neox.layers.24.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
280 |
+
"gpt_neox.layers.25.attention.bias": "pytorch_model-00002-of-00002.bin",
|
281 |
+
"gpt_neox.layers.25.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
|
282 |
+
"gpt_neox.layers.25.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
|
283 |
+
"gpt_neox.layers.25.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
|
284 |
+
"gpt_neox.layers.25.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
|
285 |
+
"gpt_neox.layers.25.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
|
286 |
+
"gpt_neox.layers.25.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
|
287 |
+
"gpt_neox.layers.25.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
288 |
+
"gpt_neox.layers.25.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
289 |
+
"gpt_neox.layers.25.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
|
290 |
+
"gpt_neox.layers.25.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
|
291 |
+
"gpt_neox.layers.25.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
|
292 |
+
"gpt_neox.layers.25.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
|
293 |
+
"gpt_neox.layers.25.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
294 |
+
"gpt_neox.layers.25.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
295 |
+
"gpt_neox.layers.26.attention.bias": "pytorch_model-00002-of-00002.bin",
|
296 |
+
"gpt_neox.layers.26.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
|
297 |
+
"gpt_neox.layers.26.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
|
298 |
+
"gpt_neox.layers.26.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
|
299 |
+
"gpt_neox.layers.26.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
|
300 |
+
"gpt_neox.layers.26.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
|
301 |
+
"gpt_neox.layers.26.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
|
302 |
+
"gpt_neox.layers.26.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
303 |
+
"gpt_neox.layers.26.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
304 |
+
"gpt_neox.layers.26.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
|
305 |
+
"gpt_neox.layers.26.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
|
306 |
+
"gpt_neox.layers.26.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
|
307 |
+
"gpt_neox.layers.26.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
|
308 |
+
"gpt_neox.layers.26.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
309 |
+
"gpt_neox.layers.26.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
310 |
+
"gpt_neox.layers.27.attention.bias": "pytorch_model-00002-of-00002.bin",
|
311 |
+
"gpt_neox.layers.27.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
|
312 |
+
"gpt_neox.layers.27.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
|
313 |
+
"gpt_neox.layers.27.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
|
314 |
+
"gpt_neox.layers.27.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
|
315 |
+
"gpt_neox.layers.27.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
|
316 |
+
"gpt_neox.layers.27.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
|
317 |
+
"gpt_neox.layers.27.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
318 |
+
"gpt_neox.layers.27.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
319 |
+
"gpt_neox.layers.27.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
|
320 |
+
"gpt_neox.layers.27.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
|
321 |
+
"gpt_neox.layers.27.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
|
322 |
+
"gpt_neox.layers.27.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
|
323 |
+
"gpt_neox.layers.27.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
324 |
+
"gpt_neox.layers.27.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
325 |
+
"gpt_neox.layers.28.attention.bias": "pytorch_model-00002-of-00002.bin",
|
326 |
+
"gpt_neox.layers.28.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
|
327 |
+
"gpt_neox.layers.28.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
|
328 |
+
"gpt_neox.layers.28.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
|
329 |
+
"gpt_neox.layers.28.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
|
330 |
+
"gpt_neox.layers.28.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
|
331 |
+
"gpt_neox.layers.28.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
|
332 |
+
"gpt_neox.layers.28.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
333 |
+
"gpt_neox.layers.28.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
334 |
+
"gpt_neox.layers.28.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
|
335 |
+
"gpt_neox.layers.28.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
|
336 |
+
"gpt_neox.layers.28.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
|
337 |
+
"gpt_neox.layers.28.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
|
338 |
+
"gpt_neox.layers.28.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
339 |
+
"gpt_neox.layers.28.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
340 |
+
"gpt_neox.layers.29.attention.bias": "pytorch_model-00002-of-00002.bin",
|
341 |
+
"gpt_neox.layers.29.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
|
342 |
+
"gpt_neox.layers.29.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
|
343 |
+
"gpt_neox.layers.29.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
|
344 |
+
"gpt_neox.layers.29.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
|
345 |
+
"gpt_neox.layers.29.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
|
346 |
+
"gpt_neox.layers.29.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
|
347 |
+
"gpt_neox.layers.29.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
348 |
+
"gpt_neox.layers.29.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
349 |
+
"gpt_neox.layers.29.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
|
350 |
+
"gpt_neox.layers.29.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
|
351 |
+
"gpt_neox.layers.29.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
|
352 |
+
"gpt_neox.layers.29.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
|
353 |
+
"gpt_neox.layers.29.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
354 |
+
"gpt_neox.layers.29.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
355 |
+
"gpt_neox.layers.3.attention.bias": "pytorch_model-00001-of-00002.bin",
|
356 |
+
"gpt_neox.layers.3.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
357 |
+
"gpt_neox.layers.3.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
358 |
+
"gpt_neox.layers.3.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
359 |
+
"gpt_neox.layers.3.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
360 |
+
"gpt_neox.layers.3.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
361 |
+
"gpt_neox.layers.3.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
362 |
+
"gpt_neox.layers.3.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
363 |
+
"gpt_neox.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
364 |
+
"gpt_neox.layers.3.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
365 |
+
"gpt_neox.layers.3.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
366 |
+
"gpt_neox.layers.3.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
367 |
+
"gpt_neox.layers.3.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
368 |
+
"gpt_neox.layers.3.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
369 |
+
"gpt_neox.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
370 |
+
"gpt_neox.layers.30.attention.bias": "pytorch_model-00002-of-00002.bin",
|
371 |
+
"gpt_neox.layers.30.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
|
372 |
+
"gpt_neox.layers.30.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
|
373 |
+
"gpt_neox.layers.30.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
|
374 |
+
"gpt_neox.layers.30.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
|
375 |
+
"gpt_neox.layers.30.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
|
376 |
+
"gpt_neox.layers.30.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
|
377 |
+
"gpt_neox.layers.30.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
378 |
+
"gpt_neox.layers.30.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
379 |
+
"gpt_neox.layers.30.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
|
380 |
+
"gpt_neox.layers.30.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
|
381 |
+
"gpt_neox.layers.30.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
|
382 |
+
"gpt_neox.layers.30.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
|
383 |
+
"gpt_neox.layers.30.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
384 |
+
"gpt_neox.layers.30.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
385 |
+
"gpt_neox.layers.31.attention.bias": "pytorch_model-00002-of-00002.bin",
|
386 |
+
"gpt_neox.layers.31.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
|
387 |
+
"gpt_neox.layers.31.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
|
388 |
+
"gpt_neox.layers.31.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
|
389 |
+
"gpt_neox.layers.31.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
|
390 |
+
"gpt_neox.layers.31.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
|
391 |
+
"gpt_neox.layers.31.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
|
392 |
+
"gpt_neox.layers.31.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
393 |
+
"gpt_neox.layers.31.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
394 |
+
"gpt_neox.layers.31.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
|
395 |
+
"gpt_neox.layers.31.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
|
396 |
+
"gpt_neox.layers.31.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
|
397 |
+
"gpt_neox.layers.31.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
|
398 |
+
"gpt_neox.layers.31.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
|
399 |
+
"gpt_neox.layers.31.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
|
400 |
+
"gpt_neox.layers.4.attention.bias": "pytorch_model-00001-of-00002.bin",
|
401 |
+
"gpt_neox.layers.4.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
402 |
+
"gpt_neox.layers.4.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
403 |
+
"gpt_neox.layers.4.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
404 |
+
"gpt_neox.layers.4.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
405 |
+
"gpt_neox.layers.4.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
406 |
+
"gpt_neox.layers.4.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
407 |
+
"gpt_neox.layers.4.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
408 |
+
"gpt_neox.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
409 |
+
"gpt_neox.layers.4.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
410 |
+
"gpt_neox.layers.4.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
411 |
+
"gpt_neox.layers.4.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
412 |
+
"gpt_neox.layers.4.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
413 |
+
"gpt_neox.layers.4.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
414 |
+
"gpt_neox.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
415 |
+
"gpt_neox.layers.5.attention.bias": "pytorch_model-00001-of-00002.bin",
|
416 |
+
"gpt_neox.layers.5.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
417 |
+
"gpt_neox.layers.5.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
418 |
+
"gpt_neox.layers.5.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
419 |
+
"gpt_neox.layers.5.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
420 |
+
"gpt_neox.layers.5.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
421 |
+
"gpt_neox.layers.5.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
422 |
+
"gpt_neox.layers.5.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
423 |
+
"gpt_neox.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
424 |
+
"gpt_neox.layers.5.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
425 |
+
"gpt_neox.layers.5.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
426 |
+
"gpt_neox.layers.5.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
427 |
+
"gpt_neox.layers.5.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
428 |
+
"gpt_neox.layers.5.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
429 |
+
"gpt_neox.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
430 |
+
"gpt_neox.layers.6.attention.bias": "pytorch_model-00001-of-00002.bin",
|
431 |
+
"gpt_neox.layers.6.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
432 |
+
"gpt_neox.layers.6.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
433 |
+
"gpt_neox.layers.6.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
434 |
+
"gpt_neox.layers.6.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
435 |
+
"gpt_neox.layers.6.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
436 |
+
"gpt_neox.layers.6.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
437 |
+
"gpt_neox.layers.6.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
438 |
+
"gpt_neox.layers.6.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
439 |
+
"gpt_neox.layers.6.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
440 |
+
"gpt_neox.layers.6.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
441 |
+
"gpt_neox.layers.6.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
442 |
+
"gpt_neox.layers.6.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
443 |
+
"gpt_neox.layers.6.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
444 |
+
"gpt_neox.layers.6.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
445 |
+
"gpt_neox.layers.7.attention.bias": "pytorch_model-00001-of-00002.bin",
|
446 |
+
"gpt_neox.layers.7.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
447 |
+
"gpt_neox.layers.7.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
448 |
+
"gpt_neox.layers.7.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
449 |
+
"gpt_neox.layers.7.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
450 |
+
"gpt_neox.layers.7.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
451 |
+
"gpt_neox.layers.7.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
452 |
+
"gpt_neox.layers.7.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
453 |
+
"gpt_neox.layers.7.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
454 |
+
"gpt_neox.layers.7.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
455 |
+
"gpt_neox.layers.7.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
456 |
+
"gpt_neox.layers.7.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
457 |
+
"gpt_neox.layers.7.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
458 |
+
"gpt_neox.layers.7.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
459 |
+
"gpt_neox.layers.7.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
460 |
+
"gpt_neox.layers.8.attention.bias": "pytorch_model-00001-of-00002.bin",
|
461 |
+
"gpt_neox.layers.8.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
462 |
+
"gpt_neox.layers.8.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
463 |
+
"gpt_neox.layers.8.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
464 |
+
"gpt_neox.layers.8.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
465 |
+
"gpt_neox.layers.8.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
466 |
+
"gpt_neox.layers.8.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
467 |
+
"gpt_neox.layers.8.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
468 |
+
"gpt_neox.layers.8.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
469 |
+
"gpt_neox.layers.8.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
470 |
+
"gpt_neox.layers.8.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
471 |
+
"gpt_neox.layers.8.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
472 |
+
"gpt_neox.layers.8.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
473 |
+
"gpt_neox.layers.8.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
474 |
+
"gpt_neox.layers.8.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
475 |
+
"gpt_neox.layers.9.attention.bias": "pytorch_model-00001-of-00002.bin",
|
476 |
+
"gpt_neox.layers.9.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
|
477 |
+
"gpt_neox.layers.9.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
|
478 |
+
"gpt_neox.layers.9.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
|
479 |
+
"gpt_neox.layers.9.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
|
480 |
+
"gpt_neox.layers.9.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
|
481 |
+
"gpt_neox.layers.9.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
|
482 |
+
"gpt_neox.layers.9.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
483 |
+
"gpt_neox.layers.9.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
|
484 |
+
"gpt_neox.layers.9.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
|
485 |
+
"gpt_neox.layers.9.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
|
486 |
+
"gpt_neox.layers.9.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
|
487 |
+
"gpt_neox.layers.9.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
|
488 |
+
"gpt_neox.layers.9.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
|
489 |
+
"gpt_neox.layers.9.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin"
|
490 |
+
}
|
491 |
+
}
|
special_tokens_map.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|endoftext|>",
|
3 |
+
"eos_token": "<|endoftext|>",
|
4 |
+
"unk_token": "<|endoftext|>"
|
5 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"bos_token": "<|endoftext|>",
|
4 |
+
"eos_token": "<|endoftext|>",
|
5 |
+
"name_or_path": "EleutherAI/pythia-6.9b-deduped",
|
6 |
+
"special_tokens_map_file": "/fsx/home-hailey/.cache/huggingface/hub/models--EleutherAI--gpt-neox-20b/snapshots/3523781c8df75f7741687a4284f6f70e1afa12f4/special_tokens_map.json",
|
7 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
8 |
+
"unk_token": "<|endoftext|>"
|
9 |
+
}
|