Commit
•
7e52c5b
1
Parent(s):
63d42ec
Upload 11 files
Browse files- checkpoint-1000/README.md +204 -0
- checkpoint-1000/adapter_config.json +26 -0
- checkpoint-1000/adapter_model.safetensors +3 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/rng_state.pth +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/special_tokens_map.json +24 -0
- checkpoint-1000/tokenizer.model +3 -0
- checkpoint-1000/tokenizer_config.json +45 -0
- checkpoint-1000/trainer_state.json +1421 -0
- checkpoint-1000/training_args.bin +3 -0
checkpoint-1000/README.md
ADDED
@@ -0,0 +1,204 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: ../Mixtral-8x7B-Instruct-v0.1
|
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 0.7.1
|
checkpoint-1000/adapter_config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "../Mixtral-8x7B-Instruct-v0.1",
|
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": 16.0,
|
13 |
+
"lora_dropout": 0.1,
|
14 |
+
"megatron_config": null,
|
15 |
+
"megatron_core": "megatron.core",
|
16 |
+
"modules_to_save": null,
|
17 |
+
"peft_type": "LORA",
|
18 |
+
"r": 8,
|
19 |
+
"rank_pattern": {},
|
20 |
+
"revision": null,
|
21 |
+
"target_modules": [
|
22 |
+
"q_proj",
|
23 |
+
"v_proj"
|
24 |
+
],
|
25 |
+
"task_type": "CAUSAL_LM"
|
26 |
+
}
|
checkpoint-1000/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ac138ef31736b2d07c02760fcf41354cf4d457606aac941b0bd70eeb6e77d5e9
|
3 |
+
size 13648432
|
checkpoint-1000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:73f9564433e92a1dd391462c5c9125abd9fe1dc9651d62d2c51d52fceb56db43
|
3 |
+
size 27370618
|
checkpoint-1000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
3 |
+
size 14244
|
checkpoint-1000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8de747a419ad75415fd169016ec6667a5cd4c6f32ae3f0bea715e43fc4572125
|
3 |
+
size 1064
|
checkpoint-1000/special_tokens_map.json
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"eos_token": {
|
10 |
+
"content": "</s>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": "</s>",
|
17 |
+
"unk_token": {
|
18 |
+
"content": "<unk>",
|
19 |
+
"lstrip": false,
|
20 |
+
"normalized": false,
|
21 |
+
"rstrip": false,
|
22 |
+
"single_word": false
|
23 |
+
}
|
24 |
+
}
|
checkpoint-1000/tokenizer.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
3 |
+
size 493443
|
checkpoint-1000/tokenizer_config.json
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": true,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"added_tokens_decoder": {
|
5 |
+
"0": {
|
6 |
+
"content": "<unk>",
|
7 |
+
"lstrip": false,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false,
|
11 |
+
"special": true
|
12 |
+
},
|
13 |
+
"1": {
|
14 |
+
"content": "<s>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false,
|
19 |
+
"special": true
|
20 |
+
},
|
21 |
+
"2": {
|
22 |
+
"content": "</s>",
|
23 |
+
"lstrip": false,
|
24 |
+
"normalized": false,
|
25 |
+
"rstrip": false,
|
26 |
+
"single_word": false,
|
27 |
+
"special": true
|
28 |
+
}
|
29 |
+
},
|
30 |
+
"additional_special_tokens": [],
|
31 |
+
"bos_token": "<s>",
|
32 |
+
"chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
|
33 |
+
"clean_up_tokenization_spaces": false,
|
34 |
+
"eos_token": "</s>",
|
35 |
+
"legacy": true,
|
36 |
+
"model_max_length": 1000000000000000019884624838656,
|
37 |
+
"pad_token": "</s>",
|
38 |
+
"padding_side": "right",
|
39 |
+
"sp_model_kwargs": {},
|
40 |
+
"spaces_between_special_tokens": false,
|
41 |
+
"split_special_tokens": false,
|
42 |
+
"tokenizer_class": "LlamaTokenizer",
|
43 |
+
"unk_token": "<unk>",
|
44 |
+
"use_default_system_prompt": false
|
45 |
+
}
|
checkpoint-1000/trainer_state.json
ADDED
@@ -0,0 +1,1421 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.10976647183117917,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0,
|
13 |
+
"learning_rate": 9.999970269475589e-06,
|
14 |
+
"logits/chosen": -2.748720645904541,
|
15 |
+
"logits/rejected": -3.5176024436950684,
|
16 |
+
"logps/chosen": -165.87062072753906,
|
17 |
+
"logps/rejected": -23.556869506835938,
|
18 |
+
"loss": 1.5949,
|
19 |
+
"rewards/accuracies": 0.612500011920929,
|
20 |
+
"rewards/chosen": 11.860421180725098,
|
21 |
+
"rewards/margins": 2.147804021835327,
|
22 |
+
"rewards/rejected": 9.712616920471191,
|
23 |
+
"step": 10
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.0,
|
27 |
+
"learning_rate": 9.999881078255916e-06,
|
28 |
+
"logits/chosen": -3.261986255645752,
|
29 |
+
"logits/rejected": -4.041715145111084,
|
30 |
+
"logps/chosen": -153.11898803710938,
|
31 |
+
"logps/rejected": -24.845947265625,
|
32 |
+
"loss": 1.6208,
|
33 |
+
"rewards/accuracies": 0.574999988079071,
|
34 |
+
"rewards/chosen": 12.161478042602539,
|
35 |
+
"rewards/margins": 2.482882499694824,
|
36 |
+
"rewards/rejected": 9.678595542907715,
|
37 |
+
"step": 20
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.0,
|
41 |
+
"learning_rate": 9.99973242740166e-06,
|
42 |
+
"logits/chosen": -2.687460422515869,
|
43 |
+
"logits/rejected": -3.837062358856201,
|
44 |
+
"logps/chosen": -148.68341064453125,
|
45 |
+
"logps/rejected": -34.57600402832031,
|
46 |
+
"loss": 1.497,
|
47 |
+
"rewards/accuracies": 0.699999988079071,
|
48 |
+
"rewards/chosen": 12.840588569641113,
|
49 |
+
"rewards/margins": 2.2458953857421875,
|
50 |
+
"rewards/rejected": 10.59469223022461,
|
51 |
+
"step": 30
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.0,
|
55 |
+
"learning_rate": 9.99954780509173e-06,
|
56 |
+
"logits/chosen": -2.9765782356262207,
|
57 |
+
"logits/rejected": -4.164349555969238,
|
58 |
+
"logps/chosen": -201.6632537841797,
|
59 |
+
"logps/rejected": -29.7652645111084,
|
60 |
+
"loss": 1.8816,
|
61 |
+
"rewards/accuracies": 0.574999988079071,
|
62 |
+
"rewards/chosen": 12.654102325439453,
|
63 |
+
"rewards/margins": 2.569594383239746,
|
64 |
+
"rewards/rejected": 10.084507942199707,
|
65 |
+
"step": 40
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.01,
|
69 |
+
"learning_rate": 9.999286186386431e-06,
|
70 |
+
"logits/chosen": -3.153574228286743,
|
71 |
+
"logits/rejected": -3.934051036834717,
|
72 |
+
"logps/chosen": -138.03778076171875,
|
73 |
+
"logps/rejected": -30.519153594970703,
|
74 |
+
"loss": 1.9103,
|
75 |
+
"rewards/accuracies": 0.6499999761581421,
|
76 |
+
"rewards/chosen": 11.428102493286133,
|
77 |
+
"rewards/margins": 1.7032209634780884,
|
78 |
+
"rewards/rejected": 9.724882125854492,
|
79 |
+
"step": 50
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.01,
|
83 |
+
"learning_rate": 9.998965115121132e-06,
|
84 |
+
"logits/chosen": -3.201958417892456,
|
85 |
+
"logits/rejected": -4.274596691131592,
|
86 |
+
"logps/chosen": -166.93260192871094,
|
87 |
+
"logps/rejected": -35.59614944458008,
|
88 |
+
"loss": 2.0164,
|
89 |
+
"rewards/accuracies": 0.5625,
|
90 |
+
"rewards/chosen": 11.013410568237305,
|
91 |
+
"rewards/margins": 1.4899368286132812,
|
92 |
+
"rewards/rejected": 9.523473739624023,
|
93 |
+
"step": 60
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.01,
|
97 |
+
"learning_rate": 9.998584595114074e-06,
|
98 |
+
"logits/chosen": -3.6508994102478027,
|
99 |
+
"logits/rejected": -4.988117694854736,
|
100 |
+
"logps/chosen": -192.28738403320312,
|
101 |
+
"logps/rejected": -39.88438415527344,
|
102 |
+
"loss": 0.7081,
|
103 |
+
"rewards/accuracies": 0.7875000238418579,
|
104 |
+
"rewards/chosen": 12.313508987426758,
|
105 |
+
"rewards/margins": 4.361058712005615,
|
106 |
+
"rewards/rejected": 7.952448844909668,
|
107 |
+
"step": 70
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.01,
|
111 |
+
"learning_rate": 9.998144630890485e-06,
|
112 |
+
"logits/chosen": -3.2785956859588623,
|
113 |
+
"logits/rejected": -4.609368801116943,
|
114 |
+
"logps/chosen": -177.5666961669922,
|
115 |
+
"logps/rejected": -48.05603790283203,
|
116 |
+
"loss": 1.1081,
|
117 |
+
"rewards/accuracies": 0.699999988079071,
|
118 |
+
"rewards/chosen": 12.321931838989258,
|
119 |
+
"rewards/margins": 4.14510440826416,
|
120 |
+
"rewards/rejected": 8.176828384399414,
|
121 |
+
"step": 80
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.01,
|
125 |
+
"learning_rate": 9.997645227682513e-06,
|
126 |
+
"logits/chosen": -3.4411368370056152,
|
127 |
+
"logits/rejected": -4.750131607055664,
|
128 |
+
"logps/chosen": -163.21450805664062,
|
129 |
+
"logps/rejected": -53.2952880859375,
|
130 |
+
"loss": 1.1417,
|
131 |
+
"rewards/accuracies": 0.737500011920929,
|
132 |
+
"rewards/chosen": 11.179890632629395,
|
133 |
+
"rewards/margins": 3.490216016769409,
|
134 |
+
"rewards/rejected": 7.689675331115723,
|
135 |
+
"step": 90
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.01,
|
139 |
+
"learning_rate": 9.997086391429161e-06,
|
140 |
+
"logits/chosen": -3.429443836212158,
|
141 |
+
"logits/rejected": -4.5238776206970215,
|
142 |
+
"logps/chosen": -176.43289184570312,
|
143 |
+
"logps/rejected": -65.92268371582031,
|
144 |
+
"loss": 0.9445,
|
145 |
+
"rewards/accuracies": 0.7250000238418579,
|
146 |
+
"rewards/chosen": 10.390297889709473,
|
147 |
+
"rewards/margins": 3.6286094188690186,
|
148 |
+
"rewards/rejected": 6.761689186096191,
|
149 |
+
"step": 100
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.01,
|
153 |
+
"learning_rate": 9.99646812877623e-06,
|
154 |
+
"logits/chosen": -3.4572105407714844,
|
155 |
+
"logits/rejected": -4.460841655731201,
|
156 |
+
"logps/chosen": -170.5079803466797,
|
157 |
+
"logps/rejected": -68.17359161376953,
|
158 |
+
"loss": 0.6313,
|
159 |
+
"rewards/accuracies": 0.8125,
|
160 |
+
"rewards/chosen": 10.23057746887207,
|
161 |
+
"rewards/margins": 4.633530616760254,
|
162 |
+
"rewards/rejected": 5.5970458984375,
|
163 |
+
"step": 110
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.01,
|
167 |
+
"learning_rate": 9.995790447076232e-06,
|
168 |
+
"logits/chosen": -3.5251846313476562,
|
169 |
+
"logits/rejected": -4.237984657287598,
|
170 |
+
"logps/chosen": -163.4308624267578,
|
171 |
+
"logps/rejected": -69.18915557861328,
|
172 |
+
"loss": 0.8116,
|
173 |
+
"rewards/accuracies": 0.824999988079071,
|
174 |
+
"rewards/chosen": 10.12580394744873,
|
175 |
+
"rewards/margins": 4.935164451599121,
|
176 |
+
"rewards/rejected": 5.190640449523926,
|
177 |
+
"step": 120
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.01,
|
181 |
+
"learning_rate": 9.995053354388295e-06,
|
182 |
+
"logits/chosen": -3.2857518196105957,
|
183 |
+
"logits/rejected": -4.0882792472839355,
|
184 |
+
"logps/chosen": -211.8911590576172,
|
185 |
+
"logps/rejected": -76.1227035522461,
|
186 |
+
"loss": 0.1832,
|
187 |
+
"rewards/accuracies": 0.8999999761581421,
|
188 |
+
"rewards/chosen": 10.317492485046387,
|
189 |
+
"rewards/margins": 5.812389373779297,
|
190 |
+
"rewards/rejected": 4.505103588104248,
|
191 |
+
"step": 130
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.02,
|
195 |
+
"learning_rate": 9.994256859478082e-06,
|
196 |
+
"logits/chosen": -3.0444934368133545,
|
197 |
+
"logits/rejected": -3.676633358001709,
|
198 |
+
"logps/chosen": -229.4088134765625,
|
199 |
+
"logps/rejected": -82.24313354492188,
|
200 |
+
"loss": 0.3931,
|
201 |
+
"rewards/accuracies": 0.862500011920929,
|
202 |
+
"rewards/chosen": 10.056417465209961,
|
203 |
+
"rewards/margins": 5.8769307136535645,
|
204 |
+
"rewards/rejected": 4.17948579788208,
|
205 |
+
"step": 140
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.02,
|
209 |
+
"learning_rate": 9.993489232972927e-06,
|
210 |
+
"logits/chosen": -3.0874276161193848,
|
211 |
+
"logits/rejected": -3.6212425231933594,
|
212 |
+
"logps/chosen": -163.163330078125,
|
213 |
+
"logps/rejected": -76.16889953613281,
|
214 |
+
"loss": 0.7483,
|
215 |
+
"rewards/accuracies": 0.800000011920929,
|
216 |
+
"rewards/chosen": 9.392515182495117,
|
217 |
+
"rewards/margins": 5.074815273284912,
|
218 |
+
"rewards/rejected": 4.3177008628845215,
|
219 |
+
"step": 150
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.02,
|
223 |
+
"learning_rate": 9.992579900519751e-06,
|
224 |
+
"logits/chosen": -2.9442169666290283,
|
225 |
+
"logits/rejected": -3.488318681716919,
|
226 |
+
"logps/chosen": -202.21347045898438,
|
227 |
+
"logps/rejected": -84.56549072265625,
|
228 |
+
"loss": 0.3173,
|
229 |
+
"rewards/accuracies": 0.9125000238418579,
|
230 |
+
"rewards/chosen": 9.955198287963867,
|
231 |
+
"rewards/margins": 6.719505310058594,
|
232 |
+
"rewards/rejected": 3.2356934547424316,
|
233 |
+
"step": 160
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.02,
|
237 |
+
"learning_rate": 9.99161119525913e-06,
|
238 |
+
"logits/chosen": -2.819887161254883,
|
239 |
+
"logits/rejected": -3.2742202281951904,
|
240 |
+
"logps/chosen": -212.2724151611328,
|
241 |
+
"logps/rejected": -90.73429870605469,
|
242 |
+
"loss": 0.3464,
|
243 |
+
"rewards/accuracies": 0.875,
|
244 |
+
"rewards/chosen": 9.708670616149902,
|
245 |
+
"rewards/margins": 5.832002639770508,
|
246 |
+
"rewards/rejected": 3.8766684532165527,
|
247 |
+
"step": 170
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.02,
|
251 |
+
"learning_rate": 9.990688606280938e-06,
|
252 |
+
"logits/chosen": -2.7344448566436768,
|
253 |
+
"logits/rejected": -2.890084743499756,
|
254 |
+
"logps/chosen": -184.69085693359375,
|
255 |
+
"logps/rejected": -89.32521057128906,
|
256 |
+
"loss": 0.6789,
|
257 |
+
"rewards/accuracies": 0.8374999761581421,
|
258 |
+
"rewards/chosen": 9.630158424377441,
|
259 |
+
"rewards/margins": 5.920450687408447,
|
260 |
+
"rewards/rejected": 3.7097079753875732,
|
261 |
+
"step": 180
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.02,
|
265 |
+
"learning_rate": 9.989607125007361e-06,
|
266 |
+
"logits/chosen": -2.5483996868133545,
|
267 |
+
"logits/rejected": -2.572554588317871,
|
268 |
+
"logps/chosen": -172.837158203125,
|
269 |
+
"logps/rejected": -92.3455581665039,
|
270 |
+
"loss": 0.4471,
|
271 |
+
"rewards/accuracies": 0.824999988079071,
|
272 |
+
"rewards/chosen": 8.161615371704102,
|
273 |
+
"rewards/margins": 4.883408069610596,
|
274 |
+
"rewards/rejected": 3.2782065868377686,
|
275 |
+
"step": 190
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.02,
|
279 |
+
"learning_rate": 9.98846630627921e-06,
|
280 |
+
"logits/chosen": -2.7279348373413086,
|
281 |
+
"logits/rejected": -2.8250045776367188,
|
282 |
+
"logps/chosen": -216.2030487060547,
|
283 |
+
"logps/rejected": -94.36392211914062,
|
284 |
+
"loss": 0.1686,
|
285 |
+
"rewards/accuracies": 0.925000011920929,
|
286 |
+
"rewards/chosen": 9.2470703125,
|
287 |
+
"rewards/margins": 6.067578315734863,
|
288 |
+
"rewards/rejected": 3.1794912815093994,
|
289 |
+
"step": 200
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.02,
|
293 |
+
"learning_rate": 9.98726616366334e-06,
|
294 |
+
"logits/chosen": -2.476562261581421,
|
295 |
+
"logits/rejected": -2.579338550567627,
|
296 |
+
"logps/chosen": -169.55120849609375,
|
297 |
+
"logps/rejected": -101.41313171386719,
|
298 |
+
"loss": 0.3145,
|
299 |
+
"rewards/accuracies": 0.875,
|
300 |
+
"rewards/chosen": 8.830245018005371,
|
301 |
+
"rewards/margins": 5.8404154777526855,
|
302 |
+
"rewards/rejected": 2.9898295402526855,
|
303 |
+
"step": 210
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.02,
|
307 |
+
"learning_rate": 9.986006711432099e-06,
|
308 |
+
"logits/chosen": -2.6179986000061035,
|
309 |
+
"logits/rejected": -2.938459873199463,
|
310 |
+
"logps/chosen": -239.5026397705078,
|
311 |
+
"logps/rejected": -93.572998046875,
|
312 |
+
"loss": 0.136,
|
313 |
+
"rewards/accuracies": 0.9125000238418579,
|
314 |
+
"rewards/chosen": 10.406990051269531,
|
315 |
+
"rewards/margins": 7.794189453125,
|
316 |
+
"rewards/rejected": 2.612799882888794,
|
317 |
+
"step": 220
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.03,
|
321 |
+
"learning_rate": 9.984687964563159e-06,
|
322 |
+
"logits/chosen": -2.4739646911621094,
|
323 |
+
"logits/rejected": -2.5754473209381104,
|
324 |
+
"logps/chosen": -178.31881713867188,
|
325 |
+
"logps/rejected": -95.78547668457031,
|
326 |
+
"loss": 0.2827,
|
327 |
+
"rewards/accuracies": 0.9624999761581421,
|
328 |
+
"rewards/chosen": 8.740915298461914,
|
329 |
+
"rewards/margins": 6.064324855804443,
|
330 |
+
"rewards/rejected": 2.676591396331787,
|
331 |
+
"step": 230
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.03,
|
335 |
+
"learning_rate": 9.983309938739332e-06,
|
336 |
+
"logits/chosen": -2.6331839561462402,
|
337 |
+
"logits/rejected": -2.8605971336364746,
|
338 |
+
"logps/chosen": -229.9063720703125,
|
339 |
+
"logps/rejected": -90.2779541015625,
|
340 |
+
"loss": 0.0817,
|
341 |
+
"rewards/accuracies": 0.9750000238418579,
|
342 |
+
"rewards/chosen": 10.207468032836914,
|
343 |
+
"rewards/margins": 7.982739448547363,
|
344 |
+
"rewards/rejected": 2.2247281074523926,
|
345 |
+
"step": 240
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.03,
|
349 |
+
"learning_rate": 9.98187265034839e-06,
|
350 |
+
"logits/chosen": -2.3521595001220703,
|
351 |
+
"logits/rejected": -2.5626819133758545,
|
352 |
+
"logps/chosen": -190.8962860107422,
|
353 |
+
"logps/rejected": -104.4652099609375,
|
354 |
+
"loss": 0.2826,
|
355 |
+
"rewards/accuracies": 0.8999999761581421,
|
356 |
+
"rewards/chosen": 8.663321495056152,
|
357 |
+
"rewards/margins": 6.1150994300842285,
|
358 |
+
"rewards/rejected": 2.5482218265533447,
|
359 |
+
"step": 250
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.03,
|
363 |
+
"learning_rate": 9.980376116482871e-06,
|
364 |
+
"logits/chosen": -2.4073798656463623,
|
365 |
+
"logits/rejected": -2.531911849975586,
|
366 |
+
"logps/chosen": -204.7512664794922,
|
367 |
+
"logps/rejected": -108.78132629394531,
|
368 |
+
"loss": 0.2884,
|
369 |
+
"rewards/accuracies": 0.925000011920929,
|
370 |
+
"rewards/chosen": 9.377767562866211,
|
371 |
+
"rewards/margins": 6.906778812408447,
|
372 |
+
"rewards/rejected": 2.470989227294922,
|
373 |
+
"step": 260
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.03,
|
377 |
+
"learning_rate": 9.978820354939866e-06,
|
378 |
+
"logits/chosen": -2.2742762565612793,
|
379 |
+
"logits/rejected": -2.1533379554748535,
|
380 |
+
"logps/chosen": -189.8150177001953,
|
381 |
+
"logps/rejected": -112.7627944946289,
|
382 |
+
"loss": 0.4643,
|
383 |
+
"rewards/accuracies": 0.8999999761581421,
|
384 |
+
"rewards/chosen": 9.204684257507324,
|
385 |
+
"rewards/margins": 6.9484758377075195,
|
386 |
+
"rewards/rejected": 2.2562096118927,
|
387 |
+
"step": 270
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.03,
|
391 |
+
"learning_rate": 9.97720538422082e-06,
|
392 |
+
"logits/chosen": -2.1974568367004395,
|
393 |
+
"logits/rejected": -2.233710765838623,
|
394 |
+
"logps/chosen": -198.45361328125,
|
395 |
+
"logps/rejected": -111.978515625,
|
396 |
+
"loss": 0.1247,
|
397 |
+
"rewards/accuracies": 0.9750000238418579,
|
398 |
+
"rewards/chosen": 10.231064796447754,
|
399 |
+
"rewards/margins": 8.133554458618164,
|
400 |
+
"rewards/rejected": 2.0975089073181152,
|
401 |
+
"step": 280
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.03,
|
405 |
+
"learning_rate": 9.975531223531301e-06,
|
406 |
+
"logits/chosen": -2.2490475177764893,
|
407 |
+
"logits/rejected": -2.30739164352417,
|
408 |
+
"logps/chosen": -187.04519653320312,
|
409 |
+
"logps/rejected": -109.82148742675781,
|
410 |
+
"loss": 0.1268,
|
411 |
+
"rewards/accuracies": 0.9375,
|
412 |
+
"rewards/chosen": 8.633919715881348,
|
413 |
+
"rewards/margins": 6.911574363708496,
|
414 |
+
"rewards/rejected": 1.7223446369171143,
|
415 |
+
"step": 290
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.03,
|
419 |
+
"learning_rate": 9.973797892780783e-06,
|
420 |
+
"logits/chosen": -2.035712718963623,
|
421 |
+
"logits/rejected": -1.961692452430725,
|
422 |
+
"logps/chosen": -179.4069366455078,
|
423 |
+
"logps/rejected": -116.109130859375,
|
424 |
+
"loss": 0.0938,
|
425 |
+
"rewards/accuracies": 0.9750000238418579,
|
426 |
+
"rewards/chosen": 9.934889793395996,
|
427 |
+
"rewards/margins": 8.173246383666992,
|
428 |
+
"rewards/rejected": 1.7616443634033203,
|
429 |
+
"step": 300
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.03,
|
433 |
+
"learning_rate": 9.972005412582397e-06,
|
434 |
+
"logits/chosen": -2.2436442375183105,
|
435 |
+
"logits/rejected": -1.8246514797210693,
|
436 |
+
"logps/chosen": -201.26980590820312,
|
437 |
+
"logps/rejected": -117.66944885253906,
|
438 |
+
"loss": 0.2073,
|
439 |
+
"rewards/accuracies": 0.8999999761581421,
|
440 |
+
"rewards/chosen": 8.626791954040527,
|
441 |
+
"rewards/margins": 7.419760227203369,
|
442 |
+
"rewards/rejected": 1.207031488418579,
|
443 |
+
"step": 310
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.04,
|
447 |
+
"learning_rate": 9.970153804252692e-06,
|
448 |
+
"logits/chosen": -1.8984334468841553,
|
449 |
+
"logits/rejected": -1.6126270294189453,
|
450 |
+
"logps/chosen": -185.87290954589844,
|
451 |
+
"logps/rejected": -117.4137954711914,
|
452 |
+
"loss": 0.278,
|
453 |
+
"rewards/accuracies": 0.9125000238418579,
|
454 |
+
"rewards/chosen": 8.371435165405273,
|
455 |
+
"rewards/margins": 7.151828765869141,
|
456 |
+
"rewards/rejected": 1.2196067571640015,
|
457 |
+
"step": 320
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.04,
|
461 |
+
"learning_rate": 9.968243089811384e-06,
|
462 |
+
"logits/chosen": -1.771867036819458,
|
463 |
+
"logits/rejected": -1.5103999376296997,
|
464 |
+
"logps/chosen": -192.2825469970703,
|
465 |
+
"logps/rejected": -118.695556640625,
|
466 |
+
"loss": 0.0729,
|
467 |
+
"rewards/accuracies": 0.949999988079071,
|
468 |
+
"rewards/chosen": 8.544290542602539,
|
469 |
+
"rewards/margins": 7.602395057678223,
|
470 |
+
"rewards/rejected": 0.9418962597846985,
|
471 |
+
"step": 330
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.04,
|
475 |
+
"learning_rate": 9.96627329198109e-06,
|
476 |
+
"logits/chosen": -1.924703598022461,
|
477 |
+
"logits/rejected": -1.3986445665359497,
|
478 |
+
"logps/chosen": -203.59628295898438,
|
479 |
+
"logps/rejected": -128.64163208007812,
|
480 |
+
"loss": 0.2888,
|
481 |
+
"rewards/accuracies": 0.925000011920929,
|
482 |
+
"rewards/chosen": 8.449407577514648,
|
483 |
+
"rewards/margins": 7.650280952453613,
|
484 |
+
"rewards/rejected": 0.7991273999214172,
|
485 |
+
"step": 340
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.04,
|
489 |
+
"learning_rate": 9.96424443418706e-06,
|
490 |
+
"logits/chosen": -1.8718311786651611,
|
491 |
+
"logits/rejected": -1.552980661392212,
|
492 |
+
"logps/chosen": -181.55087280273438,
|
493 |
+
"logps/rejected": -120.4705581665039,
|
494 |
+
"loss": 0.0762,
|
495 |
+
"rewards/accuracies": 0.9750000238418579,
|
496 |
+
"rewards/chosen": 8.385054588317871,
|
497 |
+
"rewards/margins": 7.822892665863037,
|
498 |
+
"rewards/rejected": 0.5621613264083862,
|
499 |
+
"step": 350
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.04,
|
503 |
+
"learning_rate": 9.962156540556895e-06,
|
504 |
+
"logits/chosen": -1.7886946201324463,
|
505 |
+
"logits/rejected": -1.5114587545394897,
|
506 |
+
"logps/chosen": -183.34066772460938,
|
507 |
+
"logps/rejected": -117.0723648071289,
|
508 |
+
"loss": 0.1247,
|
509 |
+
"rewards/accuracies": 0.9624999761581421,
|
510 |
+
"rewards/chosen": 7.962329864501953,
|
511 |
+
"rewards/margins": 7.366267204284668,
|
512 |
+
"rewards/rejected": 0.5960620045661926,
|
513 |
+
"step": 360
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.04,
|
517 |
+
"learning_rate": 9.960009635920264e-06,
|
518 |
+
"logits/chosen": -1.788691520690918,
|
519 |
+
"logits/rejected": -1.3814653158187866,
|
520 |
+
"logps/chosen": -197.45651245117188,
|
521 |
+
"logps/rejected": -127.6612319946289,
|
522 |
+
"loss": 0.076,
|
523 |
+
"rewards/accuracies": 0.9750000238418579,
|
524 |
+
"rewards/chosen": 9.019963264465332,
|
525 |
+
"rewards/margins": 8.529821395874023,
|
526 |
+
"rewards/rejected": 0.49014148116111755,
|
527 |
+
"step": 370
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.04,
|
531 |
+
"learning_rate": 9.95780374580861e-06,
|
532 |
+
"logits/chosen": -1.5638988018035889,
|
533 |
+
"logits/rejected": -1.1953327655792236,
|
534 |
+
"logps/chosen": -189.52597045898438,
|
535 |
+
"logps/rejected": -122.5257797241211,
|
536 |
+
"loss": 0.2198,
|
537 |
+
"rewards/accuracies": 0.9375,
|
538 |
+
"rewards/chosen": 7.37133264541626,
|
539 |
+
"rewards/margins": 6.876678466796875,
|
540 |
+
"rewards/rejected": 0.4946538507938385,
|
541 |
+
"step": 380
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.04,
|
545 |
+
"learning_rate": 9.955538896454835e-06,
|
546 |
+
"logits/chosen": -1.5664819478988647,
|
547 |
+
"logits/rejected": -0.962997317314148,
|
548 |
+
"logps/chosen": -204.67657470703125,
|
549 |
+
"logps/rejected": -129.93968200683594,
|
550 |
+
"loss": 0.0446,
|
551 |
+
"rewards/accuracies": 0.9624999761581421,
|
552 |
+
"rewards/chosen": 7.449831485748291,
|
553 |
+
"rewards/margins": 7.785650730133057,
|
554 |
+
"rewards/rejected": -0.33581945300102234,
|
555 |
+
"step": 390
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.04,
|
559 |
+
"learning_rate": 9.95321511479301e-06,
|
560 |
+
"logits/chosen": -1.7048896551132202,
|
561 |
+
"logits/rejected": -1.1939642429351807,
|
562 |
+
"logps/chosen": -223.2345428466797,
|
563 |
+
"logps/rejected": -130.3861846923828,
|
564 |
+
"loss": 0.1745,
|
565 |
+
"rewards/accuracies": 0.949999988079071,
|
566 |
+
"rewards/chosen": 8.562383651733398,
|
567 |
+
"rewards/margins": 8.441825866699219,
|
568 |
+
"rewards/rejected": 0.12055821716785431,
|
569 |
+
"step": 400
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.05,
|
573 |
+
"learning_rate": 9.950832428458025e-06,
|
574 |
+
"logits/chosen": -1.2762507200241089,
|
575 |
+
"logits/rejected": -0.9500716924667358,
|
576 |
+
"logps/chosen": -195.8381805419922,
|
577 |
+
"logps/rejected": -124.5715560913086,
|
578 |
+
"loss": 0.1329,
|
579 |
+
"rewards/accuracies": 0.949999988079071,
|
580 |
+
"rewards/chosen": 7.099652290344238,
|
581 |
+
"rewards/margins": 7.511727809906006,
|
582 |
+
"rewards/rejected": -0.41207608580589294,
|
583 |
+
"step": 410
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.05,
|
587 |
+
"learning_rate": 9.948390865785295e-06,
|
588 |
+
"logits/chosen": -1.151587963104248,
|
589 |
+
"logits/rejected": -0.925134539604187,
|
590 |
+
"logps/chosen": -199.50967407226562,
|
591 |
+
"logps/rejected": -137.6339111328125,
|
592 |
+
"loss": 0.1983,
|
593 |
+
"rewards/accuracies": 0.9750000238418579,
|
594 |
+
"rewards/chosen": 7.476855278015137,
|
595 |
+
"rewards/margins": 7.712057590484619,
|
596 |
+
"rewards/rejected": -0.23520107567310333,
|
597 |
+
"step": 420
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.05,
|
601 |
+
"learning_rate": 9.94589045581039e-06,
|
602 |
+
"logits/chosen": -1.0321991443634033,
|
603 |
+
"logits/rejected": -0.8457671999931335,
|
604 |
+
"logps/chosen": -206.13467407226562,
|
605 |
+
"logps/rejected": -130.86544799804688,
|
606 |
+
"loss": 0.2554,
|
607 |
+
"rewards/accuracies": 0.9375,
|
608 |
+
"rewards/chosen": 7.223177433013916,
|
609 |
+
"rewards/margins": 7.446470737457275,
|
610 |
+
"rewards/rejected": -0.2232942134141922,
|
611 |
+
"step": 430
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.05,
|
615 |
+
"learning_rate": 9.943331228268711e-06,
|
616 |
+
"logits/chosen": -1.1853946447372437,
|
617 |
+
"logits/rejected": -0.7296536564826965,
|
618 |
+
"logps/chosen": -217.77304077148438,
|
619 |
+
"logps/rejected": -127.19853210449219,
|
620 |
+
"loss": 0.1477,
|
621 |
+
"rewards/accuracies": 0.9624999761581421,
|
622 |
+
"rewards/chosen": 7.660037040710449,
|
623 |
+
"rewards/margins": 8.213974952697754,
|
624 |
+
"rewards/rejected": -0.5539380311965942,
|
625 |
+
"step": 440
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.05,
|
629 |
+
"learning_rate": 9.94071321359513e-06,
|
630 |
+
"logits/chosen": -0.9670246243476868,
|
631 |
+
"logits/rejected": -0.314952552318573,
|
632 |
+
"logps/chosen": -225.93533325195312,
|
633 |
+
"logps/rejected": -154.71826171875,
|
634 |
+
"loss": 0.172,
|
635 |
+
"rewards/accuracies": 0.949999988079071,
|
636 |
+
"rewards/chosen": 7.689838409423828,
|
637 |
+
"rewards/margins": 8.839930534362793,
|
638 |
+
"rewards/rejected": -1.1500914096832275,
|
639 |
+
"step": 450
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.05,
|
643 |
+
"learning_rate": 9.938036442923624e-06,
|
644 |
+
"logits/chosen": -0.9485570788383484,
|
645 |
+
"logits/rejected": -0.33327311277389526,
|
646 |
+
"logps/chosen": -225.99398803710938,
|
647 |
+
"logps/rejected": -142.39173889160156,
|
648 |
+
"loss": 0.0579,
|
649 |
+
"rewards/accuracies": 0.9750000238418579,
|
650 |
+
"rewards/chosen": 7.543333530426025,
|
651 |
+
"rewards/margins": 8.586603164672852,
|
652 |
+
"rewards/rejected": -1.0432696342468262,
|
653 |
+
"step": 460
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.05,
|
657 |
+
"learning_rate": 9.935300948086915e-06,
|
658 |
+
"logits/chosen": -0.8986530303955078,
|
659 |
+
"logits/rejected": -0.400226354598999,
|
660 |
+
"logps/chosen": -209.4599609375,
|
661 |
+
"logps/rejected": -130.23941040039062,
|
662 |
+
"loss": 0.1114,
|
663 |
+
"rewards/accuracies": 0.9750000238418579,
|
664 |
+
"rewards/chosen": 7.468968391418457,
|
665 |
+
"rewards/margins": 8.008504867553711,
|
666 |
+
"rewards/rejected": -0.5395378470420837,
|
667 |
+
"step": 470
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.05,
|
671 |
+
"learning_rate": 9.932506761616078e-06,
|
672 |
+
"logits/chosen": -0.7118775248527527,
|
673 |
+
"logits/rejected": -0.03559774532914162,
|
674 |
+
"logps/chosen": -193.45834350585938,
|
675 |
+
"logps/rejected": -131.99844360351562,
|
676 |
+
"loss": 0.0508,
|
677 |
+
"rewards/accuracies": 0.987500011920929,
|
678 |
+
"rewards/chosen": 7.474764823913574,
|
679 |
+
"rewards/margins": 8.308823585510254,
|
680 |
+
"rewards/rejected": -0.8340586423873901,
|
681 |
+
"step": 480
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.05,
|
685 |
+
"learning_rate": 9.929653916740168e-06,
|
686 |
+
"logits/chosen": -1.1100711822509766,
|
687 |
+
"logits/rejected": -0.40355220437049866,
|
688 |
+
"logps/chosen": -242.05697631835938,
|
689 |
+
"logps/rejected": -139.188232421875,
|
690 |
+
"loss": 0.1127,
|
691 |
+
"rewards/accuracies": 0.9624999761581421,
|
692 |
+
"rewards/chosen": 8.726511001586914,
|
693 |
+
"rewards/margins": 9.716850280761719,
|
694 |
+
"rewards/rejected": -0.9903383255004883,
|
695 |
+
"step": 490
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.05,
|
699 |
+
"learning_rate": 9.92674244738581e-06,
|
700 |
+
"logits/chosen": -0.8486649394035339,
|
701 |
+
"logits/rejected": 0.05283903330564499,
|
702 |
+
"logps/chosen": -155.493408203125,
|
703 |
+
"logps/rejected": -141.4785919189453,
|
704 |
+
"loss": 0.0901,
|
705 |
+
"rewards/accuracies": 0.949999988079071,
|
706 |
+
"rewards/chosen": 7.535587310791016,
|
707 |
+
"rewards/margins": 8.244181632995605,
|
708 |
+
"rewards/rejected": -0.7085943818092346,
|
709 |
+
"step": 500
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.06,
|
713 |
+
"learning_rate": 9.923772388176812e-06,
|
714 |
+
"logits/chosen": -0.907477080821991,
|
715 |
+
"logits/rejected": -0.171137273311615,
|
716 |
+
"logps/chosen": -236.19198608398438,
|
717 |
+
"logps/rejected": -133.06643676757812,
|
718 |
+
"loss": 0.0178,
|
719 |
+
"rewards/accuracies": 0.987500011920929,
|
720 |
+
"rewards/chosen": 8.850728034973145,
|
721 |
+
"rewards/margins": 10.053266525268555,
|
722 |
+
"rewards/rejected": -1.202539086341858,
|
723 |
+
"step": 510
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.06,
|
727 |
+
"learning_rate": 9.92074377443374e-06,
|
728 |
+
"logits/chosen": -1.0009920597076416,
|
729 |
+
"logits/rejected": -0.20153608918190002,
|
730 |
+
"logps/chosen": -178.60147094726562,
|
731 |
+
"logps/rejected": -143.53695678710938,
|
732 |
+
"loss": 0.1976,
|
733 |
+
"rewards/accuracies": 0.949999988079071,
|
734 |
+
"rewards/chosen": 7.2967023849487305,
|
735 |
+
"rewards/margins": 8.339073181152344,
|
736 |
+
"rewards/rejected": -1.0423698425292969,
|
737 |
+
"step": 520
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.06,
|
741 |
+
"learning_rate": 9.917656642173505e-06,
|
742 |
+
"logits/chosen": -1.0008546113967896,
|
743 |
+
"logits/rejected": -0.01453827042132616,
|
744 |
+
"logps/chosen": -206.38461303710938,
|
745 |
+
"logps/rejected": -144.1584014892578,
|
746 |
+
"loss": 0.1575,
|
747 |
+
"rewards/accuracies": 0.9750000238418579,
|
748 |
+
"rewards/chosen": 8.916428565979004,
|
749 |
+
"rewards/margins": 9.663307189941406,
|
750 |
+
"rewards/rejected": -0.7468785047531128,
|
751 |
+
"step": 530
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.06,
|
755 |
+
"learning_rate": 9.91451102810893e-06,
|
756 |
+
"logits/chosen": -1.326416015625,
|
757 |
+
"logits/rejected": -0.38943323493003845,
|
758 |
+
"logps/chosen": -220.2677764892578,
|
759 |
+
"logps/rejected": -130.0069122314453,
|
760 |
+
"loss": 0.054,
|
761 |
+
"rewards/accuracies": 0.987500011920929,
|
762 |
+
"rewards/chosen": 8.959368705749512,
|
763 |
+
"rewards/margins": 9.750789642333984,
|
764 |
+
"rewards/rejected": -0.7914207577705383,
|
765 |
+
"step": 540
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.06,
|
769 |
+
"learning_rate": 9.911306969648318e-06,
|
770 |
+
"logits/chosen": -0.932703971862793,
|
771 |
+
"logits/rejected": -0.0587722584605217,
|
772 |
+
"logps/chosen": -205.8779296875,
|
773 |
+
"logps/rejected": -135.040283203125,
|
774 |
+
"loss": 0.1965,
|
775 |
+
"rewards/accuracies": 0.925000011920929,
|
776 |
+
"rewards/chosen": 8.557300567626953,
|
777 |
+
"rewards/margins": 9.26187515258789,
|
778 |
+
"rewards/rejected": -0.7045742869377136,
|
779 |
+
"step": 550
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.06,
|
783 |
+
"learning_rate": 9.908044504895004e-06,
|
784 |
+
"logits/chosen": -0.9326754808425903,
|
785 |
+
"logits/rejected": -0.19441956281661987,
|
786 |
+
"logps/chosen": -195.15328979492188,
|
787 |
+
"logps/rejected": -131.31869506835938,
|
788 |
+
"loss": 0.0943,
|
789 |
+
"rewards/accuracies": 0.9624999761581421,
|
790 |
+
"rewards/chosen": 8.366579055786133,
|
791 |
+
"rewards/margins": 9.570085525512695,
|
792 |
+
"rewards/rejected": -1.203505277633667,
|
793 |
+
"step": 560
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.06,
|
797 |
+
"learning_rate": 9.904723672646903e-06,
|
798 |
+
"logits/chosen": -0.9648317098617554,
|
799 |
+
"logits/rejected": -0.06580810248851776,
|
800 |
+
"logps/chosen": -211.77633666992188,
|
801 |
+
"logps/rejected": -136.65756225585938,
|
802 |
+
"loss": 0.2817,
|
803 |
+
"rewards/accuracies": 0.925000011920929,
|
804 |
+
"rewards/chosen": 8.721487045288086,
|
805 |
+
"rewards/margins": 9.342923164367676,
|
806 |
+
"rewards/rejected": -0.6214349865913391,
|
807 |
+
"step": 570
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.06,
|
811 |
+
"learning_rate": 9.901344512396049e-06,
|
812 |
+
"logits/chosen": -1.1457253694534302,
|
813 |
+
"logits/rejected": -0.2043023556470871,
|
814 |
+
"logps/chosen": -210.77450561523438,
|
815 |
+
"logps/rejected": -128.11520385742188,
|
816 |
+
"loss": 0.1145,
|
817 |
+
"rewards/accuracies": 0.9750000238418579,
|
818 |
+
"rewards/chosen": 8.162195205688477,
|
819 |
+
"rewards/margins": 9.146950721740723,
|
820 |
+
"rewards/rejected": -0.9847553968429565,
|
821 |
+
"step": 580
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.06,
|
825 |
+
"learning_rate": 9.897907064328125e-06,
|
826 |
+
"logits/chosen": -0.8582237362861633,
|
827 |
+
"logits/rejected": -0.14161944389343262,
|
828 |
+
"logps/chosen": -202.4379425048828,
|
829 |
+
"logps/rejected": -136.9176483154297,
|
830 |
+
"loss": 0.2544,
|
831 |
+
"rewards/accuracies": 0.949999988079071,
|
832 |
+
"rewards/chosen": 8.603880882263184,
|
833 |
+
"rewards/margins": 9.390780448913574,
|
834 |
+
"rewards/rejected": -0.7868987321853638,
|
835 |
+
"step": 590
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.07,
|
839 |
+
"learning_rate": 9.894411369321984e-06,
|
840 |
+
"logits/chosen": -0.9913557767868042,
|
841 |
+
"logits/rejected": -0.0038151517510414124,
|
842 |
+
"logps/chosen": -182.00146484375,
|
843 |
+
"logps/rejected": -135.45352172851562,
|
844 |
+
"loss": 0.0794,
|
845 |
+
"rewards/accuracies": 0.949999988079071,
|
846 |
+
"rewards/chosen": 7.4787445068359375,
|
847 |
+
"rewards/margins": 8.599211692810059,
|
848 |
+
"rewards/rejected": -1.120466947555542,
|
849 |
+
"step": 600
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.07,
|
853 |
+
"learning_rate": 9.890857468949164e-06,
|
854 |
+
"logits/chosen": -0.9070969820022583,
|
855 |
+
"logits/rejected": 0.03064725361764431,
|
856 |
+
"logps/chosen": -223.1108856201172,
|
857 |
+
"logps/rejected": -145.6663055419922,
|
858 |
+
"loss": 0.0421,
|
859 |
+
"rewards/accuracies": 0.9750000238418579,
|
860 |
+
"rewards/chosen": 8.449647903442383,
|
861 |
+
"rewards/margins": 10.343244552612305,
|
862 |
+
"rewards/rejected": -1.8935970067977905,
|
863 |
+
"step": 610
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.07,
|
867 |
+
"learning_rate": 9.887245405473393e-06,
|
868 |
+
"logits/chosen": -0.7473093867301941,
|
869 |
+
"logits/rejected": 0.2375013381242752,
|
870 |
+
"logps/chosen": -218.11428833007812,
|
871 |
+
"logps/rejected": -137.17213439941406,
|
872 |
+
"loss": 0.0778,
|
873 |
+
"rewards/accuracies": 0.9624999761581421,
|
874 |
+
"rewards/chosen": 7.949044704437256,
|
875 |
+
"rewards/margins": 8.823598861694336,
|
876 |
+
"rewards/rejected": -0.8745536804199219,
|
877 |
+
"step": 620
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.07,
|
881 |
+
"learning_rate": 9.88357522185009e-06,
|
882 |
+
"logits/chosen": -0.913166344165802,
|
883 |
+
"logits/rejected": -0.011747682467103004,
|
884 |
+
"logps/chosen": -220.34603881835938,
|
885 |
+
"logps/rejected": -141.53561401367188,
|
886 |
+
"loss": 0.0287,
|
887 |
+
"rewards/accuracies": 0.987500011920929,
|
888 |
+
"rewards/chosen": 8.446084976196289,
|
889 |
+
"rewards/margins": 9.644007682800293,
|
890 |
+
"rewards/rejected": -1.1979233026504517,
|
891 |
+
"step": 630
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.07,
|
895 |
+
"learning_rate": 9.879846961725846e-06,
|
896 |
+
"logits/chosen": -0.8051003217697144,
|
897 |
+
"logits/rejected": 0.030248349532485008,
|
898 |
+
"logps/chosen": -199.5961456298828,
|
899 |
+
"logps/rejected": -155.56056213378906,
|
900 |
+
"loss": 0.0765,
|
901 |
+
"rewards/accuracies": 0.949999988079071,
|
902 |
+
"rewards/chosen": 8.313097953796387,
|
903 |
+
"rewards/margins": 9.846205711364746,
|
904 |
+
"rewards/rejected": -1.5331077575683594,
|
905 |
+
"step": 640
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.07,
|
909 |
+
"learning_rate": 9.876060669437913e-06,
|
910 |
+
"logits/chosen": -0.616194486618042,
|
911 |
+
"logits/rejected": 0.2745420038700104,
|
912 |
+
"logps/chosen": -215.4296417236328,
|
913 |
+
"logps/rejected": -142.76492309570312,
|
914 |
+
"loss": 0.0848,
|
915 |
+
"rewards/accuracies": 0.9375,
|
916 |
+
"rewards/chosen": 7.84356164932251,
|
917 |
+
"rewards/margins": 9.544157981872559,
|
918 |
+
"rewards/rejected": -1.7005951404571533,
|
919 |
+
"step": 650
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.07,
|
923 |
+
"learning_rate": 9.872216390013675e-06,
|
924 |
+
"logits/chosen": -0.7901605367660522,
|
925 |
+
"logits/rejected": 0.41072148084640503,
|
926 |
+
"logps/chosen": -185.0421142578125,
|
927 |
+
"logps/rejected": -144.97398376464844,
|
928 |
+
"loss": 0.1577,
|
929 |
+
"rewards/accuracies": 0.9750000238418579,
|
930 |
+
"rewards/chosen": 7.704103946685791,
|
931 |
+
"rewards/margins": 9.240386009216309,
|
932 |
+
"rewards/rejected": -1.5362818241119385,
|
933 |
+
"step": 660
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.07,
|
937 |
+
"learning_rate": 9.868314169170108e-06,
|
938 |
+
"logits/chosen": -0.5289187431335449,
|
939 |
+
"logits/rejected": 0.397865355014801,
|
940 |
+
"logps/chosen": -219.9363555908203,
|
941 |
+
"logps/rejected": -142.4893798828125,
|
942 |
+
"loss": 0.1996,
|
943 |
+
"rewards/accuracies": 0.925000011920929,
|
944 |
+
"rewards/chosen": 7.231544494628906,
|
945 |
+
"rewards/margins": 8.826276779174805,
|
946 |
+
"rewards/rejected": -1.5947315692901611,
|
947 |
+
"step": 670
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.07,
|
951 |
+
"learning_rate": 9.86435405331324e-06,
|
952 |
+
"logits/chosen": -0.3374841809272766,
|
953 |
+
"logits/rejected": 0.4616989195346832,
|
954 |
+
"logps/chosen": -197.3474578857422,
|
955 |
+
"logps/rejected": -141.45425415039062,
|
956 |
+
"loss": 0.0638,
|
957 |
+
"rewards/accuracies": 0.987500011920929,
|
958 |
+
"rewards/chosen": 7.858680725097656,
|
959 |
+
"rewards/margins": 9.586578369140625,
|
960 |
+
"rewards/rejected": -1.7278966903686523,
|
961 |
+
"step": 680
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.08,
|
965 |
+
"learning_rate": 9.860336089537601e-06,
|
966 |
+
"logits/chosen": -0.4742531180381775,
|
967 |
+
"logits/rejected": 0.3263815939426422,
|
968 |
+
"logps/chosen": -197.79933166503906,
|
969 |
+
"logps/rejected": -145.15786743164062,
|
970 |
+
"loss": 0.1294,
|
971 |
+
"rewards/accuracies": 0.949999988079071,
|
972 |
+
"rewards/chosen": 7.016458988189697,
|
973 |
+
"rewards/margins": 8.728564262390137,
|
974 |
+
"rewards/rejected": -1.7121059894561768,
|
975 |
+
"step": 690
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.08,
|
979 |
+
"learning_rate": 9.856260325625657e-06,
|
980 |
+
"logits/chosen": -0.6684874892234802,
|
981 |
+
"logits/rejected": 0.5160067677497864,
|
982 |
+
"logps/chosen": -215.1957550048828,
|
983 |
+
"logps/rejected": -157.3702392578125,
|
984 |
+
"loss": 0.0588,
|
985 |
+
"rewards/accuracies": 0.9624999761581421,
|
986 |
+
"rewards/chosen": 6.882597923278809,
|
987 |
+
"rewards/margins": 9.006999969482422,
|
988 |
+
"rewards/rejected": -2.1244006156921387,
|
989 |
+
"step": 700
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.08,
|
993 |
+
"learning_rate": 9.85212681004725e-06,
|
994 |
+
"logits/chosen": -0.7230426073074341,
|
995 |
+
"logits/rejected": 0.48965024948120117,
|
996 |
+
"logps/chosen": -238.30581665039062,
|
997 |
+
"logps/rejected": -151.21774291992188,
|
998 |
+
"loss": 0.0211,
|
999 |
+
"rewards/accuracies": 1.0,
|
1000 |
+
"rewards/chosen": 8.455827713012695,
|
1001 |
+
"rewards/margins": 10.841313362121582,
|
1002 |
+
"rewards/rejected": -2.385484457015991,
|
1003 |
+
"step": 710
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.08,
|
1007 |
+
"learning_rate": 9.847935591959011e-06,
|
1008 |
+
"logits/chosen": -0.2026168555021286,
|
1009 |
+
"logits/rejected": 0.6747661828994751,
|
1010 |
+
"logps/chosen": -206.3802490234375,
|
1011 |
+
"logps/rejected": -141.45103454589844,
|
1012 |
+
"loss": 0.0742,
|
1013 |
+
"rewards/accuracies": 0.9624999761581421,
|
1014 |
+
"rewards/chosen": 7.532872676849365,
|
1015 |
+
"rewards/margins": 9.587210655212402,
|
1016 |
+
"rewards/rejected": -2.0543372631073,
|
1017 |
+
"step": 720
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.08,
|
1021 |
+
"learning_rate": 9.84368672120379e-06,
|
1022 |
+
"logits/chosen": -0.3312041461467743,
|
1023 |
+
"logits/rejected": 0.7338079214096069,
|
1024 |
+
"logps/chosen": -200.46630859375,
|
1025 |
+
"logps/rejected": -147.34877014160156,
|
1026 |
+
"loss": 0.1475,
|
1027 |
+
"rewards/accuracies": 0.949999988079071,
|
1028 |
+
"rewards/chosen": 7.3677077293396,
|
1029 |
+
"rewards/margins": 9.639891624450684,
|
1030 |
+
"rewards/rejected": -2.272183656692505,
|
1031 |
+
"step": 730
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.08,
|
1035 |
+
"learning_rate": 9.839380248310043e-06,
|
1036 |
+
"logits/chosen": -0.16066689789295197,
|
1037 |
+
"logits/rejected": 0.8108816146850586,
|
1038 |
+
"logps/chosen": -175.85104370117188,
|
1039 |
+
"logps/rejected": -157.10450744628906,
|
1040 |
+
"loss": 0.0537,
|
1041 |
+
"rewards/accuracies": 0.987500011920929,
|
1042 |
+
"rewards/chosen": 7.471951961517334,
|
1043 |
+
"rewards/margins": 9.829510688781738,
|
1044 |
+
"rewards/rejected": -2.3575589656829834,
|
1045 |
+
"step": 740
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.08,
|
1049 |
+
"learning_rate": 9.835016224491251e-06,
|
1050 |
+
"logits/chosen": -0.3715950846672058,
|
1051 |
+
"logits/rejected": 0.7917531728744507,
|
1052 |
+
"logps/chosen": -183.4439697265625,
|
1053 |
+
"logps/rejected": -141.3053436279297,
|
1054 |
+
"loss": 0.0628,
|
1055 |
+
"rewards/accuracies": 0.9750000238418579,
|
1056 |
+
"rewards/chosen": 7.702794551849365,
|
1057 |
+
"rewards/margins": 9.382607460021973,
|
1058 |
+
"rewards/rejected": -1.6798127889633179,
|
1059 |
+
"step": 750
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.08,
|
1063 |
+
"learning_rate": 9.830594701645304e-06,
|
1064 |
+
"logits/chosen": -0.3017290234565735,
|
1065 |
+
"logits/rejected": 0.6022787094116211,
|
1066 |
+
"logps/chosen": -198.8234100341797,
|
1067 |
+
"logps/rejected": -144.2804412841797,
|
1068 |
+
"loss": 0.0508,
|
1069 |
+
"rewards/accuracies": 0.9750000238418579,
|
1070 |
+
"rewards/chosen": 7.64525842666626,
|
1071 |
+
"rewards/margins": 9.416332244873047,
|
1072 |
+
"rewards/rejected": -1.7710736989974976,
|
1073 |
+
"step": 760
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.08,
|
1077 |
+
"learning_rate": 9.826115732353877e-06,
|
1078 |
+
"logits/chosen": -0.36430418491363525,
|
1079 |
+
"logits/rejected": 0.5193418860435486,
|
1080 |
+
"logps/chosen": -256.91583251953125,
|
1081 |
+
"logps/rejected": -153.57138061523438,
|
1082 |
+
"loss": 0.1406,
|
1083 |
+
"rewards/accuracies": 0.949999988079071,
|
1084 |
+
"rewards/chosen": 7.545983791351318,
|
1085 |
+
"rewards/margins": 10.080999374389648,
|
1086 |
+
"rewards/rejected": -2.535015821456909,
|
1087 |
+
"step": 770
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.09,
|
1091 |
+
"learning_rate": 9.82157936988181e-06,
|
1092 |
+
"logits/chosen": -0.4545150697231293,
|
1093 |
+
"logits/rejected": 0.928754985332489,
|
1094 |
+
"logps/chosen": -220.9812469482422,
|
1095 |
+
"logps/rejected": -150.5502166748047,
|
1096 |
+
"loss": 0.0332,
|
1097 |
+
"rewards/accuracies": 0.987500011920929,
|
1098 |
+
"rewards/chosen": 7.531121730804443,
|
1099 |
+
"rewards/margins": 9.915712356567383,
|
1100 |
+
"rewards/rejected": -2.3845906257629395,
|
1101 |
+
"step": 780
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.09,
|
1105 |
+
"learning_rate": 9.816985668176482e-06,
|
1106 |
+
"logits/chosen": -0.7030404806137085,
|
1107 |
+
"logits/rejected": 0.5952990055084229,
|
1108 |
+
"logps/chosen": -231.7360382080078,
|
1109 |
+
"logps/rejected": -148.20033264160156,
|
1110 |
+
"loss": 0.152,
|
1111 |
+
"rewards/accuracies": 0.9750000238418579,
|
1112 |
+
"rewards/chosen": 8.060033798217773,
|
1113 |
+
"rewards/margins": 9.900753021240234,
|
1114 |
+
"rewards/rejected": -1.8407185077667236,
|
1115 |
+
"step": 790
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.09,
|
1119 |
+
"learning_rate": 9.812334681867152e-06,
|
1120 |
+
"logits/chosen": -0.7694862484931946,
|
1121 |
+
"logits/rejected": 0.30579477548599243,
|
1122 |
+
"logps/chosen": -182.74615478515625,
|
1123 |
+
"logps/rejected": -146.8044891357422,
|
1124 |
+
"loss": 0.106,
|
1125 |
+
"rewards/accuracies": 0.949999988079071,
|
1126 |
+
"rewards/chosen": 7.797210693359375,
|
1127 |
+
"rewards/margins": 9.230527877807617,
|
1128 |
+
"rewards/rejected": -1.4333173036575317,
|
1129 |
+
"step": 800
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.09,
|
1133 |
+
"learning_rate": 9.807626466264328e-06,
|
1134 |
+
"logits/chosen": -0.6776883006095886,
|
1135 |
+
"logits/rejected": 0.4075050354003906,
|
1136 |
+
"logps/chosen": -188.23257446289062,
|
1137 |
+
"logps/rejected": -147.1197052001953,
|
1138 |
+
"loss": 0.0302,
|
1139 |
+
"rewards/accuracies": 0.987500011920929,
|
1140 |
+
"rewards/chosen": 8.79939079284668,
|
1141 |
+
"rewards/margins": 10.872949600219727,
|
1142 |
+
"rewards/rejected": -2.073558807373047,
|
1143 |
+
"step": 810
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.09,
|
1147 |
+
"learning_rate": 9.802861077359098e-06,
|
1148 |
+
"logits/chosen": -0.6511849164962769,
|
1149 |
+
"logits/rejected": 0.4911740720272064,
|
1150 |
+
"logps/chosen": -197.52383422851562,
|
1151 |
+
"logps/rejected": -153.37123107910156,
|
1152 |
+
"loss": 0.1162,
|
1153 |
+
"rewards/accuracies": 0.9750000238418579,
|
1154 |
+
"rewards/chosen": 7.9149274826049805,
|
1155 |
+
"rewards/margins": 10.323246955871582,
|
1156 |
+
"rewards/rejected": -2.4083173274993896,
|
1157 |
+
"step": 820
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.09,
|
1161 |
+
"learning_rate": 9.798038571822463e-06,
|
1162 |
+
"logits/chosen": -0.8261006474494934,
|
1163 |
+
"logits/rejected": 0.3168261647224426,
|
1164 |
+
"logps/chosen": -200.90821838378906,
|
1165 |
+
"logps/rejected": -147.97134399414062,
|
1166 |
+
"loss": 0.0839,
|
1167 |
+
"rewards/accuracies": 0.9375,
|
1168 |
+
"rewards/chosen": 7.613385200500488,
|
1169 |
+
"rewards/margins": 9.651713371276855,
|
1170 |
+
"rewards/rejected": -2.038329601287842,
|
1171 |
+
"step": 830
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.09,
|
1175 |
+
"learning_rate": 9.793159007004675e-06,
|
1176 |
+
"logits/chosen": -0.9683536291122437,
|
1177 |
+
"logits/rejected": -0.2409909963607788,
|
1178 |
+
"logps/chosen": -231.24169921875,
|
1179 |
+
"logps/rejected": -154.0245361328125,
|
1180 |
+
"loss": 0.0287,
|
1181 |
+
"rewards/accuracies": 0.987500011920929,
|
1182 |
+
"rewards/chosen": 8.987038612365723,
|
1183 |
+
"rewards/margins": 10.66617202758789,
|
1184 |
+
"rewards/rejected": -1.6791340112686157,
|
1185 |
+
"step": 840
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.09,
|
1189 |
+
"learning_rate": 9.78822244093454e-06,
|
1190 |
+
"logits/chosen": -0.7381896376609802,
|
1191 |
+
"logits/rejected": 0.2965128421783447,
|
1192 |
+
"logps/chosen": -174.1750946044922,
|
1193 |
+
"logps/rejected": -151.99537658691406,
|
1194 |
+
"loss": 0.0919,
|
1195 |
+
"rewards/accuracies": 0.9624999761581421,
|
1196 |
+
"rewards/chosen": 7.649115085601807,
|
1197 |
+
"rewards/margins": 9.841694831848145,
|
1198 |
+
"rewards/rejected": -2.192580223083496,
|
1199 |
+
"step": 850
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.09,
|
1203 |
+
"learning_rate": 9.783228932318736e-06,
|
1204 |
+
"logits/chosen": -0.6059221625328064,
|
1205 |
+
"logits/rejected": 0.08314159512519836,
|
1206 |
+
"logps/chosen": -210.4627227783203,
|
1207 |
+
"logps/rejected": -152.2406768798828,
|
1208 |
+
"loss": 0.0535,
|
1209 |
+
"rewards/accuracies": 0.9750000238418579,
|
1210 |
+
"rewards/chosen": 7.937192440032959,
|
1211 |
+
"rewards/margins": 9.917118072509766,
|
1212 |
+
"rewards/rejected": -1.979925513267517,
|
1213 |
+
"step": 860
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.1,
|
1217 |
+
"learning_rate": 9.778178540541116e-06,
|
1218 |
+
"logits/chosen": -0.730563759803772,
|
1219 |
+
"logits/rejected": 0.34863799810409546,
|
1220 |
+
"logps/chosen": -200.68209838867188,
|
1221 |
+
"logps/rejected": -151.97323608398438,
|
1222 |
+
"loss": 0.0397,
|
1223 |
+
"rewards/accuracies": 0.987500011920929,
|
1224 |
+
"rewards/chosen": 7.986664772033691,
|
1225 |
+
"rewards/margins": 10.529966354370117,
|
1226 |
+
"rewards/rejected": -2.543302059173584,
|
1227 |
+
"step": 870
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.1,
|
1231 |
+
"learning_rate": 9.773071325662e-06,
|
1232 |
+
"logits/chosen": -0.8053274154663086,
|
1233 |
+
"logits/rejected": 0.12549547851085663,
|
1234 |
+
"logps/chosen": -192.92202758789062,
|
1235 |
+
"logps/rejected": -146.79693603515625,
|
1236 |
+
"loss": 0.1526,
|
1237 |
+
"rewards/accuracies": 0.9624999761581421,
|
1238 |
+
"rewards/chosen": 8.616433143615723,
|
1239 |
+
"rewards/margins": 10.316951751708984,
|
1240 |
+
"rewards/rejected": -1.7005198001861572,
|
1241 |
+
"step": 880
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.1,
|
1245 |
+
"learning_rate": 9.767907348417458e-06,
|
1246 |
+
"logits/chosen": -0.6939282417297363,
|
1247 |
+
"logits/rejected": 0.5801184773445129,
|
1248 |
+
"logps/chosen": -237.7701873779297,
|
1249 |
+
"logps/rejected": -150.30313110351562,
|
1250 |
+
"loss": 0.2363,
|
1251 |
+
"rewards/accuracies": 0.9375,
|
1252 |
+
"rewards/chosen": 8.603517532348633,
|
1253 |
+
"rewards/margins": 10.700518608093262,
|
1254 |
+
"rewards/rejected": -2.0970005989074707,
|
1255 |
+
"step": 890
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.1,
|
1259 |
+
"learning_rate": 9.762686670218588e-06,
|
1260 |
+
"logits/chosen": -0.5126532912254333,
|
1261 |
+
"logits/rejected": 0.5091619491577148,
|
1262 |
+
"logps/chosen": -213.9130401611328,
|
1263 |
+
"logps/rejected": -155.66085815429688,
|
1264 |
+
"loss": 0.0143,
|
1265 |
+
"rewards/accuracies": 1.0,
|
1266 |
+
"rewards/chosen": 8.648975372314453,
|
1267 |
+
"rewards/margins": 11.388581275939941,
|
1268 |
+
"rewards/rejected": -2.739605665206909,
|
1269 |
+
"step": 900
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.1,
|
1273 |
+
"learning_rate": 9.757409353150793e-06,
|
1274 |
+
"logits/chosen": -0.4479140341281891,
|
1275 |
+
"logits/rejected": 0.5997704267501831,
|
1276 |
+
"logps/chosen": -206.91928100585938,
|
1277 |
+
"logps/rejected": -153.65902709960938,
|
1278 |
+
"loss": 0.0504,
|
1279 |
+
"rewards/accuracies": 0.987500011920929,
|
1280 |
+
"rewards/chosen": 6.941058158874512,
|
1281 |
+
"rewards/margins": 9.638825416564941,
|
1282 |
+
"rewards/rejected": -2.6977667808532715,
|
1283 |
+
"step": 910
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.1,
|
1287 |
+
"learning_rate": 9.752075459973034e-06,
|
1288 |
+
"logits/chosen": -0.48702773451805115,
|
1289 |
+
"logits/rejected": 0.5097901225090027,
|
1290 |
+
"logps/chosen": -253.9662322998047,
|
1291 |
+
"logps/rejected": -161.4595947265625,
|
1292 |
+
"loss": 0.0841,
|
1293 |
+
"rewards/accuracies": 0.9750000238418579,
|
1294 |
+
"rewards/chosen": 8.159878730773926,
|
1295 |
+
"rewards/margins": 10.821348190307617,
|
1296 |
+
"rewards/rejected": -2.661470890045166,
|
1297 |
+
"step": 920
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.1,
|
1301 |
+
"learning_rate": 9.746685054117086e-06,
|
1302 |
+
"logits/chosen": -0.4953214228153229,
|
1303 |
+
"logits/rejected": 0.5581221580505371,
|
1304 |
+
"logps/chosen": -248.16482543945312,
|
1305 |
+
"logps/rejected": -150.33505249023438,
|
1306 |
+
"loss": 0.0869,
|
1307 |
+
"rewards/accuracies": 0.9750000238418579,
|
1308 |
+
"rewards/chosen": 7.8457465171813965,
|
1309 |
+
"rewards/margins": 10.702252388000488,
|
1310 |
+
"rewards/rejected": -2.856506109237671,
|
1311 |
+
"step": 930
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.1,
|
1315 |
+
"learning_rate": 9.741238199686789e-06,
|
1316 |
+
"logits/chosen": -0.5145618915557861,
|
1317 |
+
"logits/rejected": 0.7150206565856934,
|
1318 |
+
"logps/chosen": -236.22683715820312,
|
1319 |
+
"logps/rejected": -149.36941528320312,
|
1320 |
+
"loss": 0.0362,
|
1321 |
+
"rewards/accuracies": 0.987500011920929,
|
1322 |
+
"rewards/chosen": 7.276123046875,
|
1323 |
+
"rewards/margins": 10.051583290100098,
|
1324 |
+
"rewards/rejected": -2.775459051132202,
|
1325 |
+
"step": 940
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.1,
|
1329 |
+
"learning_rate": 9.735734961457277e-06,
|
1330 |
+
"logits/chosen": -0.4139340817928314,
|
1331 |
+
"logits/rejected": 0.7916821241378784,
|
1332 |
+
"logps/chosen": -179.39382934570312,
|
1333 |
+
"logps/rejected": -153.31869506835938,
|
1334 |
+
"loss": 0.0393,
|
1335 |
+
"rewards/accuracies": 0.9750000238418579,
|
1336 |
+
"rewards/chosen": 7.780765533447266,
|
1337 |
+
"rewards/margins": 10.50986099243164,
|
1338 |
+
"rewards/rejected": -2.729094982147217,
|
1339 |
+
"step": 950
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.11,
|
1343 |
+
"learning_rate": 9.730175404874212e-06,
|
1344 |
+
"logits/chosen": -0.3699759542942047,
|
1345 |
+
"logits/rejected": 0.7061504125595093,
|
1346 |
+
"logps/chosen": -173.6160888671875,
|
1347 |
+
"logps/rejected": -157.6312713623047,
|
1348 |
+
"loss": 0.15,
|
1349 |
+
"rewards/accuracies": 0.9375,
|
1350 |
+
"rewards/chosen": 8.363532066345215,
|
1351 |
+
"rewards/margins": 10.816950798034668,
|
1352 |
+
"rewards/rejected": -2.4534192085266113,
|
1353 |
+
"step": 960
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.11,
|
1357 |
+
"learning_rate": 9.724559596053009e-06,
|
1358 |
+
"logits/chosen": -0.6346268057823181,
|
1359 |
+
"logits/rejected": 0.48736244440078735,
|
1360 |
+
"logps/chosen": -197.9464111328125,
|
1361 |
+
"logps/rejected": -163.78814697265625,
|
1362 |
+
"loss": 0.0861,
|
1363 |
+
"rewards/accuracies": 0.9624999761581421,
|
1364 |
+
"rewards/chosen": 8.08003044128418,
|
1365 |
+
"rewards/margins": 10.763525009155273,
|
1366 |
+
"rewards/rejected": -2.683495044708252,
|
1367 |
+
"step": 970
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.11,
|
1371 |
+
"learning_rate": 9.718887601778044e-06,
|
1372 |
+
"logits/chosen": -0.4699893891811371,
|
1373 |
+
"logits/rejected": 0.5934313535690308,
|
1374 |
+
"logps/chosen": -172.5706787109375,
|
1375 |
+
"logps/rejected": -147.60012817382812,
|
1376 |
+
"loss": 0.0598,
|
1377 |
+
"rewards/accuracies": 0.9750000238418579,
|
1378 |
+
"rewards/chosen": 7.218860626220703,
|
1379 |
+
"rewards/margins": 9.611445426940918,
|
1380 |
+
"rewards/rejected": -2.3925843238830566,
|
1381 |
+
"step": 980
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.11,
|
1385 |
+
"learning_rate": 9.713159489501863e-06,
|
1386 |
+
"logits/chosen": -0.7212098836898804,
|
1387 |
+
"logits/rejected": 0.6830633282661438,
|
1388 |
+
"logps/chosen": -195.26705932617188,
|
1389 |
+
"logps/rejected": -144.74806213378906,
|
1390 |
+
"loss": 0.0301,
|
1391 |
+
"rewards/accuracies": 0.987500011920929,
|
1392 |
+
"rewards/chosen": 7.268791198730469,
|
1393 |
+
"rewards/margins": 9.785104751586914,
|
1394 |
+
"rewards/rejected": -2.5163135528564453,
|
1395 |
+
"step": 990
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.11,
|
1399 |
+
"learning_rate": 9.707956263849591e-06,
|
1400 |
+
"logits/chosen": -0.653510570526123,
|
1401 |
+
"logits/rejected": 0.6825904250144958,
|
1402 |
+
"logps/chosen": -214.06729125976562,
|
1403 |
+
"logps/rejected": -151.3599395751953,
|
1404 |
+
"loss": 0.1073,
|
1405 |
+
"rewards/accuracies": 0.9624999761581421,
|
1406 |
+
"rewards/chosen": 8.91938591003418,
|
1407 |
+
"rewards/margins": 11.606144905090332,
|
1408 |
+
"rewards/rejected": -2.6867592334747314,
|
1409 |
+
"step": 1000
|
1410 |
+
}
|
1411 |
+
],
|
1412 |
+
"logging_steps": 10,
|
1413 |
+
"max_steps": 9110,
|
1414 |
+
"num_input_tokens_seen": 0,
|
1415 |
+
"num_train_epochs": 1,
|
1416 |
+
"save_steps": 1000,
|
1417 |
+
"total_flos": 1.2405468070721618e+18,
|
1418 |
+
"train_batch_size": 2,
|
1419 |
+
"trial_name": null,
|
1420 |
+
"trial_params": null
|
1421 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:014e8566ef08057d5965d10b2d2a273d4f18ec2f8a1b5fb9328162a40b0d9e1e
|
3 |
+
size 4856
|