Upload folder using huggingface_hub
Browse files- README.md +40 -0
- adapter_config.json +27 -0
- adapter_model.safetensors +3 -0
- checkpoint-3974/README.md +204 -0
- checkpoint-3974/adapter_config.json +27 -0
- checkpoint-3974/adapter_model.safetensors +3 -0
- checkpoint-3974/merges.txt +0 -0
- checkpoint-3974/optimizer.pt +3 -0
- checkpoint-3974/pytorch_model.bin +3 -0
- checkpoint-3974/rng_state_0.pth +3 -0
- checkpoint-3974/rng_state_1.pth +3 -0
- checkpoint-3974/scheduler.pt +3 -0
- checkpoint-3974/special_tokens_map.json +6 -0
- checkpoint-3974/tokenizer.json +0 -0
- checkpoint-3974/tokenizer_config.json +21 -0
- checkpoint-3974/trainer_state.json +1127 -0
- checkpoint-3974/training_args.bin +3 -0
- checkpoint-3974/vocab.json +0 -0
- handler.py +32 -0
- merges.txt +0 -0
- requirements.txt +2 -0
- special_tokens_map.json +6 -0
- tokenizer.json +0 -0
- tokenizer_config.json +21 -0
- training_args.bin +3 -0
- training_params.json +47 -0
- vocab.json +0 -0
README.md
ADDED
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- autotrain
|
4 |
+
- text-generation
|
5 |
+
widget:
|
6 |
+
- text: "I love AutoTrain because "
|
7 |
+
license: other
|
8 |
+
---
|
9 |
+
|
10 |
+
# Model Trained Using AutoTrain
|
11 |
+
|
12 |
+
This model was trained using AutoTrain. For more information, please visit [AutoTrain](https://hf.co/docs/autotrain).
|
13 |
+
|
14 |
+
# Usage
|
15 |
+
|
16 |
+
```python
|
17 |
+
|
18 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
19 |
+
|
20 |
+
model_path = "PATH_TO_THIS_REPO"
|
21 |
+
|
22 |
+
tokenizer = AutoTokenizer.from_pretrained(model_path)
|
23 |
+
model = AutoModelForCausalLM.from_pretrained(
|
24 |
+
model_path,
|
25 |
+
device_map="auto",
|
26 |
+
torch_dtype='auto'
|
27 |
+
).eval()
|
28 |
+
|
29 |
+
# Prompt content: "hi"
|
30 |
+
messages = [
|
31 |
+
{"role": "user", "content": "hi"}
|
32 |
+
]
|
33 |
+
|
34 |
+
input_ids = tokenizer.apply_chat_template(conversation=messages, tokenize=True, add_generation_prompt=True, return_tensors='pt')
|
35 |
+
output_ids = model.generate(input_ids.to('cuda'))
|
36 |
+
response = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)
|
37 |
+
|
38 |
+
# Model response: "Hello! How can I assist you today?"
|
39 |
+
print(response)
|
40 |
+
```
|
adapter_config.json
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "openai-community/gpt2",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layers_pattern": null,
|
10 |
+
"layers_to_transform": null,
|
11 |
+
"loftq_config": {},
|
12 |
+
"lora_alpha": 32,
|
13 |
+
"lora_dropout": 0.05,
|
14 |
+
"megatron_config": null,
|
15 |
+
"megatron_core": "megatron.core",
|
16 |
+
"modules_to_save": null,
|
17 |
+
"peft_type": "LORA",
|
18 |
+
"r": 16,
|
19 |
+
"rank_pattern": {},
|
20 |
+
"revision": null,
|
21 |
+
"target_modules": [
|
22 |
+
"c_attn"
|
23 |
+
],
|
24 |
+
"task_type": "CAUSAL_LM",
|
25 |
+
"use_dora": false,
|
26 |
+
"use_rslora": false
|
27 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f548a4701f561175083517c8e8307d92758cdc482fa881d4cdbe6119540f5b5b
|
3 |
+
size 2362376
|
checkpoint-3974/README.md
ADDED
@@ -0,0 +1,204 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: openai-community/gpt2
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
|
201 |
+
|
202 |
+
### Framework versions
|
203 |
+
|
204 |
+
- PEFT 0.9.0
|
checkpoint-3974/adapter_config.json
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "openai-community/gpt2",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layers_pattern": null,
|
10 |
+
"layers_to_transform": null,
|
11 |
+
"loftq_config": {},
|
12 |
+
"lora_alpha": 32,
|
13 |
+
"lora_dropout": 0.05,
|
14 |
+
"megatron_config": null,
|
15 |
+
"megatron_core": "megatron.core",
|
16 |
+
"modules_to_save": null,
|
17 |
+
"peft_type": "LORA",
|
18 |
+
"r": 16,
|
19 |
+
"rank_pattern": {},
|
20 |
+
"revision": null,
|
21 |
+
"target_modules": [
|
22 |
+
"c_attn"
|
23 |
+
],
|
24 |
+
"task_type": "CAUSAL_LM",
|
25 |
+
"use_dora": false,
|
26 |
+
"use_rslora": false
|
27 |
+
}
|
checkpoint-3974/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f548a4701f561175083517c8e8307d92758cdc482fa881d4cdbe6119540f5b5b
|
3 |
+
size 2362376
|
checkpoint-3974/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-3974/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:eee1b1728e8438fe34ac5a1d4b3f905cdd824398966e3701d5b2df7154c1bff7
|
3 |
+
size 4739222
|
checkpoint-3974/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:049c26b844b79121ddd8379f7f69194e63f6fbf6aa007eeac0c66f17eebb8893
|
3 |
+
size 888
|
checkpoint-3974/rng_state_0.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a2f5efd5470f93f6ba67a6c1f9e2a6cf7039c47ad4f1d8f4059c3ecc90dc2d12
|
3 |
+
size 14512
|
checkpoint-3974/rng_state_1.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:81f1c5f2ea587f8120cbd4eb653bcd8566e2411e33003bf23b3bf0456d83bc15
|
3 |
+
size 14512
|
checkpoint-3974/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b5436738daf65b792bb8914b09ff52c37d8f1f0a855135763a7e9bf41f62ea41
|
3 |
+
size 1064
|
checkpoint-3974/special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|endoftext|>",
|
3 |
+
"eos_token": "<|endoftext|>",
|
4 |
+
"pad_token": "<|endoftext|>",
|
5 |
+
"unk_token": "<|endoftext|>"
|
6 |
+
}
|
checkpoint-3974/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-3974/tokenizer_config.json
ADDED
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"50256": {
|
5 |
+
"content": "<|endoftext|>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": true,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": true
|
11 |
+
}
|
12 |
+
},
|
13 |
+
"bos_token": "<|endoftext|>",
|
14 |
+
"chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
15 |
+
"clean_up_tokenization_spaces": true,
|
16 |
+
"eos_token": "<|endoftext|>",
|
17 |
+
"model_max_length": 1024,
|
18 |
+
"pad_token": "<|endoftext|>",
|
19 |
+
"tokenizer_class": "GPT2Tokenizer",
|
20 |
+
"unk_token": "<|endoftext|>"
|
21 |
+
}
|
checkpoint-3974/trainer_state.json
ADDED
@@ -0,0 +1,1127 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9998741980123286,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 3974,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.01,
|
13 |
+
"grad_norm": 0.3939279019832611,
|
14 |
+
"learning_rate": 1.2562814070351759e-05,
|
15 |
+
"loss": 4.4744,
|
16 |
+
"step": 25
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.01,
|
20 |
+
"grad_norm": 0.4029081165790558,
|
21 |
+
"learning_rate": 2.5125628140703518e-05,
|
22 |
+
"loss": 4.4566,
|
23 |
+
"step": 50
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.02,
|
27 |
+
"grad_norm": 0.38896021246910095,
|
28 |
+
"learning_rate": 3.768844221105528e-05,
|
29 |
+
"loss": 4.4112,
|
30 |
+
"step": 75
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.03,
|
34 |
+
"grad_norm": 0.3309621214866638,
|
35 |
+
"learning_rate": 5.0251256281407036e-05,
|
36 |
+
"loss": 4.2634,
|
37 |
+
"step": 100
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.03,
|
41 |
+
"grad_norm": 0.36627885699272156,
|
42 |
+
"learning_rate": 6.28140703517588e-05,
|
43 |
+
"loss": 4.1502,
|
44 |
+
"step": 125
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.04,
|
48 |
+
"grad_norm": 0.32366621494293213,
|
49 |
+
"learning_rate": 7.537688442211056e-05,
|
50 |
+
"loss": 3.9593,
|
51 |
+
"step": 150
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.04,
|
55 |
+
"grad_norm": 0.297305166721344,
|
56 |
+
"learning_rate": 8.793969849246232e-05,
|
57 |
+
"loss": 3.8315,
|
58 |
+
"step": 175
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.05,
|
62 |
+
"grad_norm": 0.31555816531181335,
|
63 |
+
"learning_rate": 0.00010050251256281407,
|
64 |
+
"loss": 3.8021,
|
65 |
+
"step": 200
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.06,
|
69 |
+
"grad_norm": 0.28000104427337646,
|
70 |
+
"learning_rate": 0.00011306532663316583,
|
71 |
+
"loss": 3.689,
|
72 |
+
"step": 225
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.06,
|
76 |
+
"grad_norm": 0.2804722487926483,
|
77 |
+
"learning_rate": 0.0001256281407035176,
|
78 |
+
"loss": 3.6452,
|
79 |
+
"step": 250
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.07,
|
83 |
+
"grad_norm": 0.3214639723300934,
|
84 |
+
"learning_rate": 0.00013819095477386937,
|
85 |
+
"loss": 3.6317,
|
86 |
+
"step": 275
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.08,
|
90 |
+
"grad_norm": 0.3243967592716217,
|
91 |
+
"learning_rate": 0.00015075376884422112,
|
92 |
+
"loss": 3.6302,
|
93 |
+
"step": 300
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.08,
|
97 |
+
"grad_norm": 0.3477385938167572,
|
98 |
+
"learning_rate": 0.00016331658291457289,
|
99 |
+
"loss": 3.5881,
|
100 |
+
"step": 325
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.09,
|
104 |
+
"grad_norm": 0.4255552589893341,
|
105 |
+
"learning_rate": 0.00017587939698492463,
|
106 |
+
"loss": 3.6165,
|
107 |
+
"step": 350
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.09,
|
111 |
+
"grad_norm": 0.32802626490592957,
|
112 |
+
"learning_rate": 0.0001884422110552764,
|
113 |
+
"loss": 3.6214,
|
114 |
+
"step": 375
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.1,
|
118 |
+
"grad_norm": 0.3549612760543823,
|
119 |
+
"learning_rate": 0.0001998881431767338,
|
120 |
+
"loss": 3.5841,
|
121 |
+
"step": 400
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.11,
|
125 |
+
"grad_norm": 0.31946343183517456,
|
126 |
+
"learning_rate": 0.00019848993288590605,
|
127 |
+
"loss": 3.544,
|
128 |
+
"step": 425
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.11,
|
132 |
+
"grad_norm": 0.38572484254837036,
|
133 |
+
"learning_rate": 0.0001970917225950783,
|
134 |
+
"loss": 3.5418,
|
135 |
+
"step": 450
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.12,
|
139 |
+
"grad_norm": 0.42264020442962646,
|
140 |
+
"learning_rate": 0.00019569351230425057,
|
141 |
+
"loss": 3.5095,
|
142 |
+
"step": 475
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.13,
|
146 |
+
"grad_norm": 0.3461235463619232,
|
147 |
+
"learning_rate": 0.00019429530201342282,
|
148 |
+
"loss": 3.5163,
|
149 |
+
"step": 500
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.13,
|
153 |
+
"grad_norm": 0.3953052759170532,
|
154 |
+
"learning_rate": 0.0001928970917225951,
|
155 |
+
"loss": 3.5208,
|
156 |
+
"step": 525
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.14,
|
160 |
+
"grad_norm": 0.36267855763435364,
|
161 |
+
"learning_rate": 0.00019149888143176735,
|
162 |
+
"loss": 3.5165,
|
163 |
+
"step": 550
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.14,
|
167 |
+
"grad_norm": 0.37477219104766846,
|
168 |
+
"learning_rate": 0.0001901006711409396,
|
169 |
+
"loss": 3.4831,
|
170 |
+
"step": 575
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.15,
|
174 |
+
"grad_norm": 0.4207465946674347,
|
175 |
+
"learning_rate": 0.00018870246085011188,
|
176 |
+
"loss": 3.4899,
|
177 |
+
"step": 600
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.16,
|
181 |
+
"grad_norm": 0.38875311613082886,
|
182 |
+
"learning_rate": 0.00018730425055928413,
|
183 |
+
"loss": 3.5172,
|
184 |
+
"step": 625
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.16,
|
188 |
+
"grad_norm": 0.37320446968078613,
|
189 |
+
"learning_rate": 0.00018590604026845638,
|
190 |
+
"loss": 3.5124,
|
191 |
+
"step": 650
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.17,
|
195 |
+
"grad_norm": 0.37162256240844727,
|
196 |
+
"learning_rate": 0.00018450782997762866,
|
197 |
+
"loss": 3.4748,
|
198 |
+
"step": 675
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.18,
|
202 |
+
"grad_norm": 0.38408517837524414,
|
203 |
+
"learning_rate": 0.00018310961968680088,
|
204 |
+
"loss": 3.486,
|
205 |
+
"step": 700
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.18,
|
209 |
+
"grad_norm": 0.3887028694152832,
|
210 |
+
"learning_rate": 0.00018171140939597316,
|
211 |
+
"loss": 3.4736,
|
212 |
+
"step": 725
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.19,
|
216 |
+
"grad_norm": 0.4443490505218506,
|
217 |
+
"learning_rate": 0.0001803131991051454,
|
218 |
+
"loss": 3.479,
|
219 |
+
"step": 750
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.19,
|
223 |
+
"grad_norm": 0.39757394790649414,
|
224 |
+
"learning_rate": 0.0001789149888143177,
|
225 |
+
"loss": 3.4492,
|
226 |
+
"step": 775
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.2,
|
230 |
+
"grad_norm": 0.35315626859664917,
|
231 |
+
"learning_rate": 0.00017751677852348994,
|
232 |
+
"loss": 3.4435,
|
233 |
+
"step": 800
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.21,
|
237 |
+
"grad_norm": 0.3870226740837097,
|
238 |
+
"learning_rate": 0.0001761185682326622,
|
239 |
+
"loss": 3.4377,
|
240 |
+
"step": 825
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.21,
|
244 |
+
"grad_norm": 0.4453731179237366,
|
245 |
+
"learning_rate": 0.00017472035794183446,
|
246 |
+
"loss": 3.4614,
|
247 |
+
"step": 850
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.22,
|
251 |
+
"grad_norm": 0.43639248609542847,
|
252 |
+
"learning_rate": 0.00017332214765100671,
|
253 |
+
"loss": 3.4889,
|
254 |
+
"step": 875
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.23,
|
258 |
+
"grad_norm": 0.42209935188293457,
|
259 |
+
"learning_rate": 0.00017192393736017897,
|
260 |
+
"loss": 3.4426,
|
261 |
+
"step": 900
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.23,
|
265 |
+
"grad_norm": 0.4586125314235687,
|
266 |
+
"learning_rate": 0.00017052572706935124,
|
267 |
+
"loss": 3.4278,
|
268 |
+
"step": 925
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.24,
|
272 |
+
"grad_norm": 0.38476869463920593,
|
273 |
+
"learning_rate": 0.0001691275167785235,
|
274 |
+
"loss": 3.4459,
|
275 |
+
"step": 950
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.25,
|
279 |
+
"grad_norm": 0.38706693053245544,
|
280 |
+
"learning_rate": 0.00016772930648769577,
|
281 |
+
"loss": 3.4564,
|
282 |
+
"step": 975
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.25,
|
286 |
+
"grad_norm": 0.3744584321975708,
|
287 |
+
"learning_rate": 0.00016633109619686802,
|
288 |
+
"loss": 3.4613,
|
289 |
+
"step": 1000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.26,
|
293 |
+
"grad_norm": 0.3903864622116089,
|
294 |
+
"learning_rate": 0.00016493288590604027,
|
295 |
+
"loss": 3.45,
|
296 |
+
"step": 1025
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.26,
|
300 |
+
"grad_norm": 0.4096889793872833,
|
301 |
+
"learning_rate": 0.00016353467561521255,
|
302 |
+
"loss": 3.405,
|
303 |
+
"step": 1050
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.27,
|
307 |
+
"grad_norm": 0.39897987246513367,
|
308 |
+
"learning_rate": 0.0001621364653243848,
|
309 |
+
"loss": 3.405,
|
310 |
+
"step": 1075
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.28,
|
314 |
+
"grad_norm": 0.39934077858924866,
|
315 |
+
"learning_rate": 0.00016073825503355705,
|
316 |
+
"loss": 3.4276,
|
317 |
+
"step": 1100
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.28,
|
321 |
+
"grad_norm": 0.38111913204193115,
|
322 |
+
"learning_rate": 0.00015934004474272933,
|
323 |
+
"loss": 3.407,
|
324 |
+
"step": 1125
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.29,
|
328 |
+
"grad_norm": 0.43283891677856445,
|
329 |
+
"learning_rate": 0.00015794183445190158,
|
330 |
+
"loss": 3.3952,
|
331 |
+
"step": 1150
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.3,
|
335 |
+
"grad_norm": 0.43224990367889404,
|
336 |
+
"learning_rate": 0.00015654362416107385,
|
337 |
+
"loss": 3.4114,
|
338 |
+
"step": 1175
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.3,
|
342 |
+
"grad_norm": 0.3882638216018677,
|
343 |
+
"learning_rate": 0.0001551454138702461,
|
344 |
+
"loss": 3.4266,
|
345 |
+
"step": 1200
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.31,
|
349 |
+
"grad_norm": 0.4062541127204895,
|
350 |
+
"learning_rate": 0.00015374720357941835,
|
351 |
+
"loss": 3.4472,
|
352 |
+
"step": 1225
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.31,
|
356 |
+
"grad_norm": 0.40274298191070557,
|
357 |
+
"learning_rate": 0.0001523489932885906,
|
358 |
+
"loss": 3.4207,
|
359 |
+
"step": 1250
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.32,
|
363 |
+
"grad_norm": 0.3987756669521332,
|
364 |
+
"learning_rate": 0.00015095078299776286,
|
365 |
+
"loss": 3.3942,
|
366 |
+
"step": 1275
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.33,
|
370 |
+
"grad_norm": 0.37407705187797546,
|
371 |
+
"learning_rate": 0.00014955257270693513,
|
372 |
+
"loss": 3.3894,
|
373 |
+
"step": 1300
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.33,
|
377 |
+
"grad_norm": 0.38793492317199707,
|
378 |
+
"learning_rate": 0.00014815436241610738,
|
379 |
+
"loss": 3.3811,
|
380 |
+
"step": 1325
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.34,
|
384 |
+
"grad_norm": 0.39553505182266235,
|
385 |
+
"learning_rate": 0.00014675615212527963,
|
386 |
+
"loss": 3.3722,
|
387 |
+
"step": 1350
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.35,
|
391 |
+
"grad_norm": 0.40836596488952637,
|
392 |
+
"learning_rate": 0.0001453579418344519,
|
393 |
+
"loss": 3.4253,
|
394 |
+
"step": 1375
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.35,
|
398 |
+
"grad_norm": 0.41444289684295654,
|
399 |
+
"learning_rate": 0.00014395973154362416,
|
400 |
+
"loss": 3.3802,
|
401 |
+
"step": 1400
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.36,
|
405 |
+
"grad_norm": 0.3889537453651428,
|
406 |
+
"learning_rate": 0.00014256152125279644,
|
407 |
+
"loss": 3.3584,
|
408 |
+
"step": 1425
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.36,
|
412 |
+
"grad_norm": 0.47589465975761414,
|
413 |
+
"learning_rate": 0.0001411633109619687,
|
414 |
+
"loss": 3.3628,
|
415 |
+
"step": 1450
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.37,
|
419 |
+
"grad_norm": 0.4026196002960205,
|
420 |
+
"learning_rate": 0.00013976510067114094,
|
421 |
+
"loss": 3.3638,
|
422 |
+
"step": 1475
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.38,
|
426 |
+
"grad_norm": 0.4240950345993042,
|
427 |
+
"learning_rate": 0.00013836689038031322,
|
428 |
+
"loss": 3.3838,
|
429 |
+
"step": 1500
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.38,
|
433 |
+
"grad_norm": 0.3979630470275879,
|
434 |
+
"learning_rate": 0.00013696868008948547,
|
435 |
+
"loss": 3.365,
|
436 |
+
"step": 1525
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.39,
|
440 |
+
"grad_norm": 0.40804243087768555,
|
441 |
+
"learning_rate": 0.00013557046979865772,
|
442 |
+
"loss": 3.3546,
|
443 |
+
"step": 1550
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.4,
|
447 |
+
"grad_norm": 0.4044090807437897,
|
448 |
+
"learning_rate": 0.00013417225950783,
|
449 |
+
"loss": 3.371,
|
450 |
+
"step": 1575
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.4,
|
454 |
+
"grad_norm": 0.4067833125591278,
|
455 |
+
"learning_rate": 0.00013277404921700225,
|
456 |
+
"loss": 3.3327,
|
457 |
+
"step": 1600
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.41,
|
461 |
+
"grad_norm": 0.44250789284706116,
|
462 |
+
"learning_rate": 0.00013137583892617452,
|
463 |
+
"loss": 3.3723,
|
464 |
+
"step": 1625
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.42,
|
468 |
+
"grad_norm": 0.3748351037502289,
|
469 |
+
"learning_rate": 0.00012997762863534677,
|
470 |
+
"loss": 3.356,
|
471 |
+
"step": 1650
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.42,
|
475 |
+
"grad_norm": 0.3744029700756073,
|
476 |
+
"learning_rate": 0.00012857941834451902,
|
477 |
+
"loss": 3.3758,
|
478 |
+
"step": 1675
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.43,
|
482 |
+
"grad_norm": 0.39975643157958984,
|
483 |
+
"learning_rate": 0.0001271812080536913,
|
484 |
+
"loss": 3.4049,
|
485 |
+
"step": 1700
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.43,
|
489 |
+
"grad_norm": 0.4390547573566437,
|
490 |
+
"learning_rate": 0.00012578299776286352,
|
491 |
+
"loss": 3.3334,
|
492 |
+
"step": 1725
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.44,
|
496 |
+
"grad_norm": 0.4836800694465637,
|
497 |
+
"learning_rate": 0.0001243847874720358,
|
498 |
+
"loss": 3.3636,
|
499 |
+
"step": 1750
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.45,
|
503 |
+
"grad_norm": 0.3915925920009613,
|
504 |
+
"learning_rate": 0.00012298657718120805,
|
505 |
+
"loss": 3.369,
|
506 |
+
"step": 1775
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.45,
|
510 |
+
"grad_norm": 0.3820320665836334,
|
511 |
+
"learning_rate": 0.00012158836689038032,
|
512 |
+
"loss": 3.3817,
|
513 |
+
"step": 1800
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.46,
|
517 |
+
"grad_norm": 0.4624602198600769,
|
518 |
+
"learning_rate": 0.00012019015659955257,
|
519 |
+
"loss": 3.3862,
|
520 |
+
"step": 1825
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.47,
|
524 |
+
"grad_norm": 0.42159679532051086,
|
525 |
+
"learning_rate": 0.00011879194630872483,
|
526 |
+
"loss": 3.3466,
|
527 |
+
"step": 1850
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.47,
|
531 |
+
"grad_norm": 0.4151483476161957,
|
532 |
+
"learning_rate": 0.0001173937360178971,
|
533 |
+
"loss": 3.3739,
|
534 |
+
"step": 1875
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.48,
|
538 |
+
"grad_norm": 0.45305246114730835,
|
539 |
+
"learning_rate": 0.00011599552572706936,
|
540 |
+
"loss": 3.3451,
|
541 |
+
"step": 1900
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.48,
|
545 |
+
"grad_norm": 0.4354460537433624,
|
546 |
+
"learning_rate": 0.00011459731543624161,
|
547 |
+
"loss": 3.3258,
|
548 |
+
"step": 1925
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.49,
|
552 |
+
"grad_norm": 0.4107976257801056,
|
553 |
+
"learning_rate": 0.00011319910514541387,
|
554 |
+
"loss": 3.3841,
|
555 |
+
"step": 1950
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.5,
|
559 |
+
"grad_norm": 0.4533105194568634,
|
560 |
+
"learning_rate": 0.00011180089485458614,
|
561 |
+
"loss": 3.3479,
|
562 |
+
"step": 1975
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.5,
|
566 |
+
"grad_norm": 0.454975962638855,
|
567 |
+
"learning_rate": 0.0001104026845637584,
|
568 |
+
"loss": 3.336,
|
569 |
+
"step": 2000
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.51,
|
573 |
+
"grad_norm": 0.5079920887947083,
|
574 |
+
"learning_rate": 0.00010900447427293066,
|
575 |
+
"loss": 3.327,
|
576 |
+
"step": 2025
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.52,
|
580 |
+
"grad_norm": 0.5311482548713684,
|
581 |
+
"learning_rate": 0.00010760626398210291,
|
582 |
+
"loss": 3.3706,
|
583 |
+
"step": 2050
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.52,
|
587 |
+
"grad_norm": 0.42152655124664307,
|
588 |
+
"learning_rate": 0.00010620805369127518,
|
589 |
+
"loss": 3.353,
|
590 |
+
"step": 2075
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.53,
|
594 |
+
"grad_norm": 0.40225550532341003,
|
595 |
+
"learning_rate": 0.00010480984340044744,
|
596 |
+
"loss": 3.3351,
|
597 |
+
"step": 2100
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.53,
|
601 |
+
"grad_norm": 0.4132668673992157,
|
602 |
+
"learning_rate": 0.0001034116331096197,
|
603 |
+
"loss": 3.308,
|
604 |
+
"step": 2125
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.54,
|
608 |
+
"grad_norm": 0.473305344581604,
|
609 |
+
"learning_rate": 0.00010201342281879196,
|
610 |
+
"loss": 3.3375,
|
611 |
+
"step": 2150
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.55,
|
615 |
+
"grad_norm": 0.43594586849212646,
|
616 |
+
"learning_rate": 0.00010061521252796422,
|
617 |
+
"loss": 3.3298,
|
618 |
+
"step": 2175
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.55,
|
622 |
+
"grad_norm": 0.5282984375953674,
|
623 |
+
"learning_rate": 9.921700223713647e-05,
|
624 |
+
"loss": 3.3179,
|
625 |
+
"step": 2200
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.56,
|
629 |
+
"grad_norm": 0.45690715312957764,
|
630 |
+
"learning_rate": 9.781879194630873e-05,
|
631 |
+
"loss": 3.3583,
|
632 |
+
"step": 2225
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.57,
|
636 |
+
"grad_norm": 0.4544149339199066,
|
637 |
+
"learning_rate": 9.642058165548098e-05,
|
638 |
+
"loss": 3.3033,
|
639 |
+
"step": 2250
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.57,
|
643 |
+
"grad_norm": 0.4424913823604584,
|
644 |
+
"learning_rate": 9.502237136465325e-05,
|
645 |
+
"loss": 3.3326,
|
646 |
+
"step": 2275
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.58,
|
650 |
+
"grad_norm": 0.4259510934352875,
|
651 |
+
"learning_rate": 9.362416107382551e-05,
|
652 |
+
"loss": 3.3324,
|
653 |
+
"step": 2300
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.58,
|
657 |
+
"grad_norm": 0.46472057700157166,
|
658 |
+
"learning_rate": 9.222595078299778e-05,
|
659 |
+
"loss": 3.3123,
|
660 |
+
"step": 2325
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.59,
|
664 |
+
"grad_norm": 0.4864007830619812,
|
665 |
+
"learning_rate": 9.082774049217003e-05,
|
666 |
+
"loss": 3.3349,
|
667 |
+
"step": 2350
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.6,
|
671 |
+
"grad_norm": 0.4463450312614441,
|
672 |
+
"learning_rate": 8.942953020134228e-05,
|
673 |
+
"loss": 3.2947,
|
674 |
+
"step": 2375
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.6,
|
678 |
+
"grad_norm": 0.4468083679676056,
|
679 |
+
"learning_rate": 8.803131991051454e-05,
|
680 |
+
"loss": 3.3423,
|
681 |
+
"step": 2400
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.61,
|
685 |
+
"grad_norm": 0.4376937448978424,
|
686 |
+
"learning_rate": 8.66331096196868e-05,
|
687 |
+
"loss": 3.3563,
|
688 |
+
"step": 2425
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.62,
|
692 |
+
"grad_norm": 0.43003320693969727,
|
693 |
+
"learning_rate": 8.523489932885907e-05,
|
694 |
+
"loss": 3.3341,
|
695 |
+
"step": 2450
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.62,
|
699 |
+
"grad_norm": 0.47375035285949707,
|
700 |
+
"learning_rate": 8.383668903803132e-05,
|
701 |
+
"loss": 3.3327,
|
702 |
+
"step": 2475
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.63,
|
706 |
+
"grad_norm": 0.42449191212654114,
|
707 |
+
"learning_rate": 8.243847874720358e-05,
|
708 |
+
"loss": 3.367,
|
709 |
+
"step": 2500
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.64,
|
713 |
+
"grad_norm": 0.455077201128006,
|
714 |
+
"learning_rate": 8.104026845637585e-05,
|
715 |
+
"loss": 3.314,
|
716 |
+
"step": 2525
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.64,
|
720 |
+
"grad_norm": 0.5218727588653564,
|
721 |
+
"learning_rate": 7.964205816554811e-05,
|
722 |
+
"loss": 3.3207,
|
723 |
+
"step": 2550
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.65,
|
727 |
+
"grad_norm": 0.40510672330856323,
|
728 |
+
"learning_rate": 7.824384787472036e-05,
|
729 |
+
"loss": 3.3055,
|
730 |
+
"step": 2575
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.65,
|
734 |
+
"grad_norm": 0.4328942596912384,
|
735 |
+
"learning_rate": 7.684563758389261e-05,
|
736 |
+
"loss": 3.3301,
|
737 |
+
"step": 2600
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.66,
|
741 |
+
"grad_norm": 0.42730334401130676,
|
742 |
+
"learning_rate": 7.544742729306487e-05,
|
743 |
+
"loss": 3.3455,
|
744 |
+
"step": 2625
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.67,
|
748 |
+
"grad_norm": 0.45374396443367004,
|
749 |
+
"learning_rate": 7.404921700223714e-05,
|
750 |
+
"loss": 3.3332,
|
751 |
+
"step": 2650
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.67,
|
755 |
+
"grad_norm": 0.4406868517398834,
|
756 |
+
"learning_rate": 7.26510067114094e-05,
|
757 |
+
"loss": 3.3121,
|
758 |
+
"step": 2675
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.68,
|
762 |
+
"grad_norm": 0.49322813749313354,
|
763 |
+
"learning_rate": 7.125279642058165e-05,
|
764 |
+
"loss": 3.3065,
|
765 |
+
"step": 2700
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.69,
|
769 |
+
"grad_norm": 0.48299440741539,
|
770 |
+
"learning_rate": 6.985458612975392e-05,
|
771 |
+
"loss": 3.2902,
|
772 |
+
"step": 2725
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.69,
|
776 |
+
"grad_norm": 0.45161083340644836,
|
777 |
+
"learning_rate": 6.851230425055929e-05,
|
778 |
+
"loss": 3.3131,
|
779 |
+
"step": 2750
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.7,
|
783 |
+
"grad_norm": 0.43272122740745544,
|
784 |
+
"learning_rate": 6.711409395973155e-05,
|
785 |
+
"loss": 3.3448,
|
786 |
+
"step": 2775
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.7,
|
790 |
+
"grad_norm": 0.4966789484024048,
|
791 |
+
"learning_rate": 6.571588366890381e-05,
|
792 |
+
"loss": 3.3083,
|
793 |
+
"step": 2800
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.71,
|
797 |
+
"grad_norm": 0.435917466878891,
|
798 |
+
"learning_rate": 6.431767337807607e-05,
|
799 |
+
"loss": 3.3005,
|
800 |
+
"step": 2825
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.72,
|
804 |
+
"grad_norm": 0.46452581882476807,
|
805 |
+
"learning_rate": 6.291946308724832e-05,
|
806 |
+
"loss": 3.3002,
|
807 |
+
"step": 2850
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.72,
|
811 |
+
"grad_norm": 0.426553875207901,
|
812 |
+
"learning_rate": 6.152125279642058e-05,
|
813 |
+
"loss": 3.3169,
|
814 |
+
"step": 2875
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.73,
|
818 |
+
"grad_norm": 0.46474194526672363,
|
819 |
+
"learning_rate": 6.012304250559284e-05,
|
820 |
+
"loss": 3.2887,
|
821 |
+
"step": 2900
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.74,
|
825 |
+
"grad_norm": 0.5483251214027405,
|
826 |
+
"learning_rate": 5.87248322147651e-05,
|
827 |
+
"loss": 3.2878,
|
828 |
+
"step": 2925
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.74,
|
832 |
+
"grad_norm": 0.4654993414878845,
|
833 |
+
"learning_rate": 5.7326621923937364e-05,
|
834 |
+
"loss": 3.3023,
|
835 |
+
"step": 2950
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.75,
|
839 |
+
"grad_norm": 0.43446651101112366,
|
840 |
+
"learning_rate": 5.592841163310962e-05,
|
841 |
+
"loss": 3.322,
|
842 |
+
"step": 2975
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.75,
|
846 |
+
"grad_norm": 0.44535747170448303,
|
847 |
+
"learning_rate": 5.4530201342281885e-05,
|
848 |
+
"loss": 3.3118,
|
849 |
+
"step": 3000
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.76,
|
853 |
+
"grad_norm": 0.4716418385505676,
|
854 |
+
"learning_rate": 5.313199105145414e-05,
|
855 |
+
"loss": 3.3211,
|
856 |
+
"step": 3025
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.77,
|
860 |
+
"grad_norm": 0.48301395773887634,
|
861 |
+
"learning_rate": 5.1733780760626406e-05,
|
862 |
+
"loss": 3.2996,
|
863 |
+
"step": 3050
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.77,
|
867 |
+
"grad_norm": 0.4252597987651825,
|
868 |
+
"learning_rate": 5.033557046979866e-05,
|
869 |
+
"loss": 3.3022,
|
870 |
+
"step": 3075
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.78,
|
874 |
+
"grad_norm": 0.4170271158218384,
|
875 |
+
"learning_rate": 4.893736017897092e-05,
|
876 |
+
"loss": 3.345,
|
877 |
+
"step": 3100
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.79,
|
881 |
+
"grad_norm": 0.42665281891822815,
|
882 |
+
"learning_rate": 4.753914988814318e-05,
|
883 |
+
"loss": 3.2633,
|
884 |
+
"step": 3125
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.79,
|
888 |
+
"grad_norm": 0.4138229787349701,
|
889 |
+
"learning_rate": 4.6140939597315434e-05,
|
890 |
+
"loss": 3.3365,
|
891 |
+
"step": 3150
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.8,
|
895 |
+
"grad_norm": 0.40854546427726746,
|
896 |
+
"learning_rate": 4.47427293064877e-05,
|
897 |
+
"loss": 3.3349,
|
898 |
+
"step": 3175
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.81,
|
902 |
+
"grad_norm": 0.4516899585723877,
|
903 |
+
"learning_rate": 4.3344519015659955e-05,
|
904 |
+
"loss": 3.2872,
|
905 |
+
"step": 3200
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.81,
|
909 |
+
"grad_norm": 0.436982125043869,
|
910 |
+
"learning_rate": 4.194630872483222e-05,
|
911 |
+
"loss": 3.296,
|
912 |
+
"step": 3225
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.82,
|
916 |
+
"grad_norm": 0.425311803817749,
|
917 |
+
"learning_rate": 4.0548098434004476e-05,
|
918 |
+
"loss": 3.2947,
|
919 |
+
"step": 3250
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.82,
|
923 |
+
"grad_norm": 0.45536327362060547,
|
924 |
+
"learning_rate": 3.914988814317674e-05,
|
925 |
+
"loss": 3.2968,
|
926 |
+
"step": 3275
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.83,
|
930 |
+
"grad_norm": 0.3970387279987335,
|
931 |
+
"learning_rate": 3.775167785234899e-05,
|
932 |
+
"loss": 3.2714,
|
933 |
+
"step": 3300
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.84,
|
937 |
+
"grad_norm": 0.4497933089733124,
|
938 |
+
"learning_rate": 3.6353467561521254e-05,
|
939 |
+
"loss": 3.3117,
|
940 |
+
"step": 3325
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.84,
|
944 |
+
"grad_norm": 0.46388792991638184,
|
945 |
+
"learning_rate": 3.495525727069351e-05,
|
946 |
+
"loss": 3.3066,
|
947 |
+
"step": 3350
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.85,
|
951 |
+
"grad_norm": 0.4629228115081787,
|
952 |
+
"learning_rate": 3.3557046979865775e-05,
|
953 |
+
"loss": 3.2982,
|
954 |
+
"step": 3375
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.86,
|
958 |
+
"grad_norm": 0.46318650245666504,
|
959 |
+
"learning_rate": 3.215883668903803e-05,
|
960 |
+
"loss": 3.3504,
|
961 |
+
"step": 3400
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.86,
|
965 |
+
"grad_norm": 0.43514785170555115,
|
966 |
+
"learning_rate": 3.076062639821029e-05,
|
967 |
+
"loss": 3.3057,
|
968 |
+
"step": 3425
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.87,
|
972 |
+
"grad_norm": 0.41652512550354004,
|
973 |
+
"learning_rate": 2.936241610738255e-05,
|
974 |
+
"loss": 3.2861,
|
975 |
+
"step": 3450
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.87,
|
979 |
+
"grad_norm": 0.4423655569553375,
|
980 |
+
"learning_rate": 2.796420581655481e-05,
|
981 |
+
"loss": 3.2521,
|
982 |
+
"step": 3475
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.88,
|
986 |
+
"grad_norm": 0.4741450846195221,
|
987 |
+
"learning_rate": 2.656599552572707e-05,
|
988 |
+
"loss": 3.3239,
|
989 |
+
"step": 3500
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.89,
|
993 |
+
"grad_norm": 0.4872516095638275,
|
994 |
+
"learning_rate": 2.516778523489933e-05,
|
995 |
+
"loss": 3.2751,
|
996 |
+
"step": 3525
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.89,
|
1000 |
+
"grad_norm": 0.4238009750843048,
|
1001 |
+
"learning_rate": 2.376957494407159e-05,
|
1002 |
+
"loss": 3.2813,
|
1003 |
+
"step": 3550
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.9,
|
1007 |
+
"grad_norm": 0.4463885724544525,
|
1008 |
+
"learning_rate": 2.237136465324385e-05,
|
1009 |
+
"loss": 3.2873,
|
1010 |
+
"step": 3575
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.91,
|
1014 |
+
"grad_norm": 0.46087589859962463,
|
1015 |
+
"learning_rate": 2.097315436241611e-05,
|
1016 |
+
"loss": 3.2789,
|
1017 |
+
"step": 3600
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.91,
|
1021 |
+
"grad_norm": 0.4878619611263275,
|
1022 |
+
"learning_rate": 1.957494407158837e-05,
|
1023 |
+
"loss": 3.3153,
|
1024 |
+
"step": 3625
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.92,
|
1028 |
+
"grad_norm": 0.4896848201751709,
|
1029 |
+
"learning_rate": 1.8176733780760627e-05,
|
1030 |
+
"loss": 3.2989,
|
1031 |
+
"step": 3650
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.92,
|
1035 |
+
"grad_norm": 0.4557226002216339,
|
1036 |
+
"learning_rate": 1.6778523489932888e-05,
|
1037 |
+
"loss": 3.3095,
|
1038 |
+
"step": 3675
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.93,
|
1042 |
+
"grad_norm": 0.43755295872688293,
|
1043 |
+
"learning_rate": 1.5380313199105145e-05,
|
1044 |
+
"loss": 3.3334,
|
1045 |
+
"step": 3700
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.94,
|
1049 |
+
"grad_norm": 0.4965628683567047,
|
1050 |
+
"learning_rate": 1.3982102908277405e-05,
|
1051 |
+
"loss": 3.2972,
|
1052 |
+
"step": 3725
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.94,
|
1056 |
+
"grad_norm": 0.4985644519329071,
|
1057 |
+
"learning_rate": 1.2583892617449666e-05,
|
1058 |
+
"loss": 3.2875,
|
1059 |
+
"step": 3750
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.95,
|
1063 |
+
"grad_norm": 0.40080615878105164,
|
1064 |
+
"learning_rate": 1.1185682326621925e-05,
|
1065 |
+
"loss": 3.2693,
|
1066 |
+
"step": 3775
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.96,
|
1070 |
+
"grad_norm": 0.42183899879455566,
|
1071 |
+
"learning_rate": 9.787472035794185e-06,
|
1072 |
+
"loss": 3.321,
|
1073 |
+
"step": 3800
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.96,
|
1077 |
+
"grad_norm": 0.5553184747695923,
|
1078 |
+
"learning_rate": 8.389261744966444e-06,
|
1079 |
+
"loss": 3.298,
|
1080 |
+
"step": 3825
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.97,
|
1084 |
+
"grad_norm": 0.4350402355194092,
|
1085 |
+
"learning_rate": 6.991051454138703e-06,
|
1086 |
+
"loss": 3.2908,
|
1087 |
+
"step": 3850
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.97,
|
1091 |
+
"grad_norm": 0.4002743363380432,
|
1092 |
+
"learning_rate": 5.592841163310962e-06,
|
1093 |
+
"loss": 3.2781,
|
1094 |
+
"step": 3875
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.98,
|
1098 |
+
"grad_norm": 0.4503081142902374,
|
1099 |
+
"learning_rate": 4.194630872483222e-06,
|
1100 |
+
"loss": 3.2933,
|
1101 |
+
"step": 3900
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.99,
|
1105 |
+
"grad_norm": 0.45145291090011597,
|
1106 |
+
"learning_rate": 2.796420581655481e-06,
|
1107 |
+
"loss": 3.2727,
|
1108 |
+
"step": 3925
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.99,
|
1112 |
+
"grad_norm": 0.42662373185157776,
|
1113 |
+
"learning_rate": 1.3982102908277406e-06,
|
1114 |
+
"loss": 3.2763,
|
1115 |
+
"step": 3950
|
1116 |
+
}
|
1117 |
+
],
|
1118 |
+
"logging_steps": 25,
|
1119 |
+
"max_steps": 3974,
|
1120 |
+
"num_input_tokens_seen": 0,
|
1121 |
+
"num_train_epochs": 1,
|
1122 |
+
"save_steps": 500,
|
1123 |
+
"total_flos": 1.6729202912919552e+16,
|
1124 |
+
"train_batch_size": 1,
|
1125 |
+
"trial_name": null,
|
1126 |
+
"trial_params": null
|
1127 |
+
}
|
checkpoint-3974/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2d7118fd57cba915713484955bbe38fb751c053c01083bbea219382833985477
|
3 |
+
size 4920
|
checkpoint-3974/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
handler.py
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from typing import Dict, List, Any
|
2 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
|
3 |
+
import torch
|
4 |
+
from peft import PeftModel
|
5 |
+
import json
|
6 |
+
import os
|
7 |
+
|
8 |
+
|
9 |
+
class EndpointHandler():
|
10 |
+
def __init__(self, path=""):
|
11 |
+
base_model_path = json.load(open(os.path.join(path, "training_params.json")))["model"]
|
12 |
+
model = AutoModelForCausalLM.from_pretrained(
|
13 |
+
base_model_path,
|
14 |
+
torch_dtype=torch.float16,
|
15 |
+
low_cpu_mem_usage=True,
|
16 |
+
trust_remote_code=True,
|
17 |
+
device_map="auto",
|
18 |
+
)
|
19 |
+
tokenizer = AutoTokenizer.from_pretrained(path, trust_remote_code=True)
|
20 |
+
model.resize_token_embeddings(len(tokenizer))
|
21 |
+
model = PeftModel.from_pretrained(model, path)
|
22 |
+
model = model.merge_and_unload()
|
23 |
+
self.pipeline = pipeline("text-generation", model=model, tokenizer=tokenizer)
|
24 |
+
|
25 |
+
def __call__(self, data: Any) -> List[List[Dict[str, float]]]:
|
26 |
+
inputs = data.pop("inputs", data)
|
27 |
+
parameters = data.pop("parameters", None)
|
28 |
+
if parameters is not None:
|
29 |
+
prediction = self.pipeline(inputs, **parameters)
|
30 |
+
else:
|
31 |
+
prediction = self.pipeline(inputs)
|
32 |
+
return prediction
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
requirements.txt
ADDED
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
1 |
+
peft==0.9.0
|
2 |
+
transformers==4.38.2
|
special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|endoftext|>",
|
3 |
+
"eos_token": "<|endoftext|>",
|
4 |
+
"pad_token": "<|endoftext|>",
|
5 |
+
"unk_token": "<|endoftext|>"
|
6 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"50256": {
|
5 |
+
"content": "<|endoftext|>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": true,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": true
|
11 |
+
}
|
12 |
+
},
|
13 |
+
"bos_token": "<|endoftext|>",
|
14 |
+
"chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
15 |
+
"clean_up_tokenization_spaces": true,
|
16 |
+
"eos_token": "<|endoftext|>",
|
17 |
+
"model_max_length": 1024,
|
18 |
+
"pad_token": "<|endoftext|>",
|
19 |
+
"tokenizer_class": "GPT2Tokenizer",
|
20 |
+
"unk_token": "<|endoftext|>"
|
21 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2d7118fd57cba915713484955bbe38fb751c053c01083bbea219382833985477
|
3 |
+
size 4920
|
training_params.json
ADDED
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"model": "openai-community/gpt2",
|
3 |
+
"project_name": "my-autotrain-llm",
|
4 |
+
"data_path": "my-autotrain-llm/autotrain-data",
|
5 |
+
"train_split": "train",
|
6 |
+
"valid_split": null,
|
7 |
+
"add_eos_token": false,
|
8 |
+
"block_size": 1024,
|
9 |
+
"model_max_length": 1024,
|
10 |
+
"padding": null,
|
11 |
+
"trainer": "default",
|
12 |
+
"use_flash_attention_2": false,
|
13 |
+
"log": "none",
|
14 |
+
"disable_gradient_checkpointing": false,
|
15 |
+
"logging_steps": -1,
|
16 |
+
"evaluation_strategy": "epoch",
|
17 |
+
"save_total_limit": 1,
|
18 |
+
"save_strategy": "epoch",
|
19 |
+
"auto_find_batch_size": false,
|
20 |
+
"mixed_precision": "fp16",
|
21 |
+
"lr": 0.0002,
|
22 |
+
"epochs": 1,
|
23 |
+
"batch_size": 1,
|
24 |
+
"warmup_ratio": 0.1,
|
25 |
+
"gradient_accumulation": 4,
|
26 |
+
"optimizer": "adamw_torch",
|
27 |
+
"scheduler": "linear",
|
28 |
+
"weight_decay": 0.01,
|
29 |
+
"max_grad_norm": 1.0,
|
30 |
+
"seed": 42,
|
31 |
+
"chat_template": null,
|
32 |
+
"quantization": "int4",
|
33 |
+
"target_modules": null,
|
34 |
+
"merge_adapter": false,
|
35 |
+
"peft": true,
|
36 |
+
"lora_r": 16,
|
37 |
+
"lora_alpha": 32,
|
38 |
+
"lora_dropout": 0.05,
|
39 |
+
"model_ref": null,
|
40 |
+
"dpo_beta": 0.1,
|
41 |
+
"prompt_text_column": "autotrain_prompt",
|
42 |
+
"text_column": "autotrain_text",
|
43 |
+
"rejected_text_column": "autotrain_rejected_text",
|
44 |
+
"push_to_hub": true,
|
45 |
+
"repo_id": "rishiai/gpt2tyyyy",
|
46 |
+
"username": null
|
47 |
+
}
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|