Model save
Browse files- README.md +29 -36
- all_results.json +4 -18
- model.safetensors +1 -1
- train_results.json +4 -4
- trainer_state.json +364 -364
README.md
CHANGED
@@ -2,16 +2,9 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: hZzy/qwen2.5-0.5b-sft-news-IFT
|
4 |
tags:
|
5 |
-
- alignment-handbook
|
6 |
-
- ndcg
|
7 |
- trl
|
8 |
- expo
|
9 |
- generated_from_trainer
|
10 |
-
- trl
|
11 |
-
- expo
|
12 |
-
- generated_from_trainer
|
13 |
-
datasets:
|
14 |
-
- hZzy/train_pairwise
|
15 |
model-index:
|
16 |
- name: qwen2.5-0.5b-expo-DPO-ES-1
|
17 |
results: []
|
@@ -20,21 +13,21 @@ model-index:
|
|
20 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
21 |
should probably proofread and complete it, then remove this comment. -->
|
22 |
|
23 |
-
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/zhiyuzha-university-of-florida/huggingface/runs/
|
24 |
# qwen2.5-0.5b-expo-DPO-ES-1
|
25 |
|
26 |
-
This model is a fine-tuned version of [hZzy/qwen2.5-0.5b-sft-news-IFT](https://huggingface.co/hZzy/qwen2.5-0.5b-sft-news-IFT) on
|
27 |
It achieves the following results on the evaluation set:
|
28 |
-
- Loss: 2.
|
29 |
-
- Logps: -83.
|
30 |
-
- Logits: -0.
|
31 |
-
- Objective: 2.
|
32 |
-
- Dpo Loss: 2.
|
33 |
-
- Regularize: 2.
|
34 |
-
- Ranking Simple: 0.
|
35 |
- Ranking Idealized: 0.5295
|
36 |
- Ranking Idealized Expo: 0.5212
|
37 |
-
- Wo Beta: 6.
|
38 |
|
39 |
## Model description
|
40 |
|
@@ -71,25 +64,25 @@ The following hyperparameters were used during training:
|
|
71 |
|
72 |
| Training Loss | Epoch | Step | Validation Loss | Logps | Logits | Objective | Dpo Loss | Regularize | Ranking Simple | Ranking Idealized | Ranking Idealized Expo | Wo Beta |
|
73 |
|:-------------:|:------:|:----:|:---------------:|:--------:|:-------:|:---------:|:--------:|:----------:|:--------------:|:-----------------:|:----------------------:|:-------:|
|
74 |
-
| 0.7017 | 0.1417 | 50 | 0.8470 | -93.
|
75 |
-
| 0.8112 | 0.2834 | 100 | 1.0529 | -86.
|
76 |
-
| 1.0895 | 0.4251 | 150 | 1.
|
77 |
-
| 1.
|
78 |
-
| 1.
|
79 |
-
| 1.1899 | 0.8503 | 300 | 2.
|
80 |
-
| 1.
|
81 |
-
| 0.725 | 1.1337 | 400 | 2.
|
82 |
-
| 0.
|
83 |
-
| 0.
|
84 |
-
| 0.
|
85 |
-
| 0.
|
86 |
-
| 0.
|
87 |
-
| 0.
|
88 |
-
| 0.
|
89 |
-
| 0.
|
90 |
-
| 0.
|
91 |
-
| 0.
|
92 |
-
| 0.
|
93 |
|
94 |
|
95 |
### Framework versions
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: hZzy/qwen2.5-0.5b-sft-news-IFT
|
4 |
tags:
|
|
|
|
|
5 |
- trl
|
6 |
- expo
|
7 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
|
|
8 |
model-index:
|
9 |
- name: qwen2.5-0.5b-expo-DPO-ES-1
|
10 |
results: []
|
|
|
13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
should probably proofread and complete it, then remove this comment. -->
|
15 |
|
16 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/zhiyuzha-university-of-florida/huggingface/runs/zba5f93y)
|
17 |
# qwen2.5-0.5b-expo-DPO-ES-1
|
18 |
|
19 |
+
This model is a fine-tuned version of [hZzy/qwen2.5-0.5b-sft-news-IFT](https://huggingface.co/hZzy/qwen2.5-0.5b-sft-news-IFT) on an unknown dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 2.3304
|
22 |
+
- Logps: -83.6421
|
23 |
+
- Logits: -0.7129
|
24 |
+
- Objective: 2.2523
|
25 |
+
- Dpo Loss: 2.2523
|
26 |
+
- Regularize: 2.2523
|
27 |
+
- Ranking Simple: 0.5336
|
28 |
- Ranking Idealized: 0.5295
|
29 |
- Ranking Idealized Expo: 0.5212
|
30 |
+
- Wo Beta: 6.8942
|
31 |
|
32 |
## Model description
|
33 |
|
|
|
64 |
|
65 |
| Training Loss | Epoch | Step | Validation Loss | Logps | Logits | Objective | Dpo Loss | Regularize | Ranking Simple | Ranking Idealized | Ranking Idealized Expo | Wo Beta |
|
66 |
|:-------------:|:------:|:----:|:---------------:|:--------:|:-------:|:---------:|:--------:|:----------:|:--------------:|:-----------------:|:----------------------:|:-------:|
|
67 |
+
| 0.7017 | 0.1417 | 50 | 0.8470 | -93.0245 | -1.4582 | 0.8570 | 0.8570 | 0.8570 | 0.5238 | 0.5295 | 0.5212 | 7.8506 |
|
68 |
+
| 0.8112 | 0.2834 | 100 | 1.0529 | -86.6804 | -1.4383 | 1.0274 | 1.0274 | 1.0274 | 0.5285 | 0.5295 | 0.5212 | 7.4982 |
|
69 |
+
| 1.0895 | 0.4251 | 150 | 1.4498 | -84.4335 | -1.2965 | 1.4010 | 1.4010 | 1.4010 | 0.5321 | 0.5295 | 0.5212 | 7.2692 |
|
70 |
+
| 1.2362 | 0.5668 | 200 | 1.7035 | -77.7194 | -1.2955 | 1.6116 | 1.6116 | 1.6116 | 0.5321 | 0.5295 | 0.5212 | 7.2264 |
|
71 |
+
| 1.3151 | 0.7085 | 250 | 1.9222 | -92.7224 | -1.2565 | 1.8319 | 1.8319 | 1.8319 | 0.5311 | 0.5295 | 0.5212 | 7.1855 |
|
72 |
+
| 1.1899 | 0.8503 | 300 | 2.0297 | -90.9351 | -0.9786 | 1.9587 | 1.9587 | 1.9587 | 0.5367 | 0.5295 | 0.5212 | 6.9337 |
|
73 |
+
| 1.1441 | 0.9920 | 350 | 2.1653 | -82.1291 | -1.0211 | 2.0545 | 2.0545 | 2.0545 | 0.5424 | 0.5295 | 0.5212 | 7.0017 |
|
74 |
+
| 0.725 | 1.1337 | 400 | 2.2886 | -84.3458 | -0.7529 | 2.2360 | 2.2360 | 2.2360 | 0.5331 | 0.5295 | 0.5212 | 7.1541 |
|
75 |
+
| 0.7626 | 1.2754 | 450 | 2.1595 | -80.5955 | -0.8863 | 2.0657 | 2.0657 | 2.0657 | 0.5326 | 0.5295 | 0.5212 | 6.7939 |
|
76 |
+
| 0.8048 | 1.4171 | 500 | 2.2134 | -82.3489 | -0.7432 | 2.0975 | 2.0975 | 2.0975 | 0.5342 | 0.5295 | 0.5212 | 6.7984 |
|
77 |
+
| 0.7106 | 1.5588 | 550 | 2.1705 | -85.0673 | -0.6665 | 2.0696 | 2.0696 | 2.0696 | 0.5321 | 0.5295 | 0.5212 | 6.8614 |
|
78 |
+
| 0.6934 | 1.7005 | 600 | 2.2127 | -81.6773 | -0.7358 | 2.0693 | 2.0693 | 2.0693 | 0.5362 | 0.5295 | 0.5212 | 6.7265 |
|
79 |
+
| 0.6885 | 1.8422 | 650 | 2.2198 | -82.8870 | -0.6787 | 2.1432 | 2.1432 | 2.1432 | 0.5362 | 0.5295 | 0.5212 | 6.8202 |
|
80 |
+
| 0.6477 | 1.9839 | 700 | 2.3420 | -83.4105 | -0.6597 | 2.2592 | 2.2592 | 2.2592 | 0.5404 | 0.5295 | 0.5212 | 6.6836 |
|
81 |
+
| 0.3785 | 2.1256 | 750 | 2.2919 | -84.0369 | -0.7841 | 2.2005 | 2.2005 | 2.2005 | 0.5435 | 0.5295 | 0.5212 | 6.8514 |
|
82 |
+
| 0.3316 | 2.2674 | 800 | 2.2220 | -84.2990 | -0.6767 | 2.1123 | 2.1123 | 2.1123 | 0.5409 | 0.5295 | 0.5212 | 6.7663 |
|
83 |
+
| 0.3283 | 2.4091 | 850 | 2.3020 | -85.0834 | -0.6538 | 2.2212 | 2.2212 | 2.2212 | 0.5409 | 0.5295 | 0.5212 | 6.7773 |
|
84 |
+
| 0.3516 | 2.5508 | 900 | 2.2723 | -84.7564 | -0.6225 | 2.1911 | 2.1911 | 2.1911 | 0.5362 | 0.5295 | 0.5212 | 6.8162 |
|
85 |
+
| 0.3245 | 2.6925 | 950 | 2.3304 | -83.6421 | -0.7129 | 2.2523 | 2.2523 | 2.2523 | 0.5336 | 0.5295 | 0.5212 | 6.8942 |
|
86 |
|
87 |
|
88 |
### Framework versions
|
all_results.json
CHANGED
@@ -1,23 +1,9 @@
|
|
1 |
{
|
2 |
"epoch": 2.69248937175248,
|
3 |
-
"eval_dpo_loss": 2.2471413612365723,
|
4 |
-
"eval_logits": -0.6650525331497192,
|
5 |
-
"eval_logps": -83.28822326660156,
|
6 |
-
"eval_loss": 2.3242855072021484,
|
7 |
-
"eval_objective": 2.2471413612365723,
|
8 |
-
"eval_ranking_idealized": 0.5295031070709229,
|
9 |
-
"eval_ranking_idealized_expo": 0.5212215185165405,
|
10 |
-
"eval_ranking_simple": 0.5377846956253052,
|
11 |
-
"eval_regularize": 2.2471413612365723,
|
12 |
-
"eval_runtime": 314.6348,
|
13 |
-
"eval_samples": 5790,
|
14 |
-
"eval_samples_per_second": 18.402,
|
15 |
-
"eval_steps_per_second": 1.535,
|
16 |
-
"eval_wo_beta": 6.681451797485352,
|
17 |
"total_flos": 0.0,
|
18 |
-
"train_loss": 0.
|
19 |
-
"train_runtime":
|
20 |
"train_samples": 50802,
|
21 |
-
"train_samples_per_second": 9.
|
22 |
-
"train_steps_per_second": 0.
|
23 |
}
|
|
|
1 |
{
|
2 |
"epoch": 2.69248937175248,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
"total_flos": 0.0,
|
4 |
+
"train_loss": 0.7491915085441188,
|
5 |
+
"train_runtime": 26152.6935,
|
6 |
"train_samples": 50802,
|
7 |
+
"train_samples_per_second": 9.713,
|
8 |
+
"train_steps_per_second": 0.067
|
9 |
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1975192208
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:92477df1ba269cbc6b373f8a1073187c1821843696650c5aab98bba461f3df44
|
3 |
size 1975192208
|
train_results.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"epoch": 2.69248937175248,
|
3 |
"total_flos": 0.0,
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
"train_samples": 50802,
|
7 |
-
"train_samples_per_second": 9.
|
8 |
-
"train_steps_per_second": 0.
|
9 |
}
|
|
|
1 |
{
|
2 |
"epoch": 2.69248937175248,
|
3 |
"total_flos": 0.0,
|
4 |
+
"train_loss": 0.7491915085441188,
|
5 |
+
"train_runtime": 26152.6935,
|
6 |
"train_samples": 50802,
|
7 |
+
"train_samples_per_second": 9.713,
|
8 |
+
"train_steps_per_second": 0.067
|
9 |
}
|
trainer_state.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"best_metric": 6.
|
3 |
"best_model_checkpoint": "./qwen2.5-0.5b/qwen2.5-0.5b-expo-DPO-ES-1/checkpoint-700",
|
4 |
"epoch": 2.69248937175248,
|
5 |
"eval_steps": 50,
|
@@ -11,7 +11,7 @@
|
|
11 |
{
|
12 |
"dpo_loss": 0.6931471824645996,
|
13 |
"epoch": 0.002834199338686821,
|
14 |
-
"grad_norm": 184.
|
15 |
"learning_rate": 2.840909090909091e-08,
|
16 |
"logits": -1.359458565711975,
|
17 |
"logps": -84.69721221923828,
|
@@ -25,640 +25,640 @@
|
|
25 |
"wo_beta": 5.271125316619873
|
26 |
},
|
27 |
{
|
28 |
-
"dpo_loss": 0.
|
29 |
"epoch": 0.14170996693434104,
|
30 |
-
"grad_norm": 184.
|
31 |
"learning_rate": 1.4204545454545458e-06,
|
32 |
-
"logits": -1.
|
33 |
-
"logps": -84.
|
34 |
"loss": 0.7017,
|
35 |
-
"objective": 0.
|
36 |
"ranking_idealized": 0.5289115905761719,
|
37 |
"ranking_idealized_expo": 0.5221088528633118,
|
38 |
"ranking_simple": 0.5225340127944946,
|
39 |
-
"regularize": 0.
|
40 |
"step": 50,
|
41 |
-
"wo_beta": 7.
|
42 |
},
|
43 |
{
|
44 |
"epoch": 0.14170996693434104,
|
45 |
-
"eval_dpo_loss": 0.
|
46 |
-
"eval_logits": -1.
|
47 |
-
"eval_logps": -93.
|
48 |
-
"eval_loss": 0.
|
49 |
-
"eval_objective": 0.
|
50 |
"eval_ranking_idealized": 0.5295031070709229,
|
51 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
52 |
"eval_ranking_simple": 0.523809552192688,
|
53 |
-
"eval_regularize": 0.
|
54 |
-
"eval_runtime":
|
55 |
-
"eval_samples_per_second": 18.
|
56 |
-
"eval_steps_per_second": 1.
|
57 |
-
"eval_wo_beta": 7.
|
58 |
"step": 50
|
59 |
},
|
60 |
{
|
61 |
-
"dpo_loss": 0.
|
62 |
"epoch": 0.2834199338686821,
|
63 |
-
"grad_norm": 165.
|
64 |
"learning_rate": 2.8409090909090916e-06,
|
65 |
-
"logits": -1.
|
66 |
-
"logps": -84.
|
67 |
"loss": 0.8112,
|
68 |
-
"objective": 0.
|
69 |
"ranking_idealized": 0.5241666436195374,
|
70 |
"ranking_idealized_expo": 0.5137500166893005,
|
71 |
"ranking_simple": 0.543749988079071,
|
72 |
-
"regularize": 0.
|
73 |
"step": 100,
|
74 |
"wo_beta": 6.3764801025390625
|
75 |
},
|
76 |
{
|
77 |
"epoch": 0.2834199338686821,
|
78 |
-
"eval_dpo_loss": 1.
|
79 |
-
"eval_logits": -1.
|
80 |
-
"eval_logps": -86.
|
81 |
-
"eval_loss": 1.
|
82 |
-
"eval_objective": 1.
|
83 |
"eval_ranking_idealized": 0.5295031070709229,
|
84 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
85 |
"eval_ranking_simple": 0.5284678936004639,
|
86 |
-
"eval_regularize": 1.
|
87 |
-
"eval_runtime":
|
88 |
-
"eval_samples_per_second":
|
89 |
-
"eval_steps_per_second": 1.
|
90 |
-
"eval_wo_beta": 7.
|
91 |
"step": 100
|
92 |
},
|
93 |
{
|
94 |
-
"dpo_loss": 1.
|
95 |
"epoch": 0.42512990080302315,
|
96 |
-
"grad_norm": 132.
|
97 |
"learning_rate": 4.2613636363636365e-06,
|
98 |
-
"logits": -1.
|
99 |
-
"logps": -75.
|
100 |
"loss": 1.0895,
|
101 |
-
"objective": 1.
|
102 |
"ranking_idealized": 0.5333333611488342,
|
103 |
"ranking_idealized_expo": 0.527916669845581,
|
104 |
"ranking_simple": 0.5645833611488342,
|
105 |
-
"regularize": 1.
|
106 |
"step": 150,
|
107 |
-
"wo_beta": 6.
|
108 |
},
|
109 |
{
|
110 |
"epoch": 0.42512990080302315,
|
111 |
-
"eval_dpo_loss": 1.
|
112 |
-
"eval_logits": -1.
|
113 |
-
"eval_logps": -84.
|
114 |
-
"eval_loss": 1.
|
115 |
-
"eval_objective": 1.
|
116 |
"eval_ranking_idealized": 0.5295031070709229,
|
117 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
118 |
"eval_ranking_simple": 0.5320910811424255,
|
119 |
-
"eval_regularize": 1.
|
120 |
-
"eval_runtime": 308.
|
121 |
-
"eval_samples_per_second": 18.
|
122 |
-
"eval_steps_per_second": 1.
|
123 |
-
"eval_wo_beta": 7.
|
124 |
"step": 150
|
125 |
},
|
126 |
{
|
127 |
-
"dpo_loss": 1.
|
128 |
"epoch": 0.5668398677373642,
|
129 |
-
"grad_norm": 120.
|
130 |
"learning_rate": 4.997168347957521e-06,
|
131 |
-
"logits": -1.
|
132 |
-
"logps": -75.
|
133 |
-
"loss": 1.
|
134 |
-
"objective": 1.
|
135 |
"ranking_idealized": 0.5204166769981384,
|
136 |
"ranking_idealized_expo": 0.51541668176651,
|
137 |
"ranking_simple": 0.5699999928474426,
|
138 |
-
"regularize": 1.
|
139 |
"step": 200,
|
140 |
-
"wo_beta": 6.
|
141 |
},
|
142 |
{
|
143 |
"epoch": 0.5668398677373642,
|
144 |
-
"eval_dpo_loss": 1.
|
145 |
-
"eval_logits": -1.
|
146 |
-
"eval_logps": -77.
|
147 |
-
"eval_loss": 1.
|
148 |
-
"eval_objective": 1.
|
149 |
"eval_ranking_idealized": 0.5295031070709229,
|
150 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
151 |
"eval_ranking_simple": 0.5320910811424255,
|
152 |
-
"eval_regularize": 1.
|
153 |
-
"eval_runtime":
|
154 |
-
"eval_samples_per_second":
|
155 |
-
"eval_steps_per_second": 1.
|
156 |
-
"eval_wo_beta": 7.
|
157 |
"step": 200
|
158 |
},
|
159 |
{
|
160 |
-
"dpo_loss": 1.
|
161 |
"epoch": 0.7085498346717053,
|
162 |
-
"grad_norm": 126.
|
163 |
"learning_rate": 4.973122855144066e-06,
|
164 |
-
"logits": -1.
|
165 |
-
"logps": -80.
|
166 |
-
"loss": 1.
|
167 |
-
"objective": 1.
|
168 |
"ranking_idealized": 0.5249999761581421,
|
169 |
"ranking_idealized_expo": 0.5162500143051147,
|
170 |
"ranking_simple": 0.5929166674613953,
|
171 |
-
"regularize": 1.
|
172 |
"step": 250,
|
173 |
-
"wo_beta": 6.
|
174 |
},
|
175 |
{
|
176 |
"epoch": 0.7085498346717053,
|
177 |
-
"eval_dpo_loss": 1.
|
178 |
-
"eval_logits": -1.
|
179 |
-
"eval_logps": -92.
|
180 |
-
"eval_loss": 1.
|
181 |
-
"eval_objective": 1.
|
182 |
"eval_ranking_idealized": 0.5295031070709229,
|
183 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
184 |
"eval_ranking_simple": 0.5310559272766113,
|
185 |
-
"eval_regularize": 1.
|
186 |
-
"eval_runtime": 308.
|
187 |
-
"eval_samples_per_second": 18.
|
188 |
-
"eval_steps_per_second": 1.
|
189 |
-
"eval_wo_beta": 7.
|
190 |
"step": 250
|
191 |
},
|
192 |
{
|
193 |
-
"dpo_loss": 1.
|
194 |
"epoch": 0.8502598016060463,
|
195 |
-
"grad_norm": 93.
|
196 |
"learning_rate": 4.924776641419513e-06,
|
197 |
-
"logits": -1.
|
198 |
-
"logps": -86.
|
199 |
"loss": 1.1899,
|
200 |
-
"objective": 1.
|
201 |
"ranking_idealized": 0.5062500238418579,
|
202 |
"ranking_idealized_expo": 0.4950000047683716,
|
203 |
"ranking_simple": 0.5833333134651184,
|
204 |
-
"regularize": 1.
|
205 |
"step": 300,
|
206 |
-
"wo_beta": 5.
|
207 |
},
|
208 |
{
|
209 |
"epoch": 0.8502598016060463,
|
210 |
-
"eval_dpo_loss": 1.
|
211 |
-
"eval_logits": -0.
|
212 |
-
"eval_logps": -90.
|
213 |
-
"eval_loss": 2.
|
214 |
-
"eval_objective": 1.
|
215 |
"eval_ranking_idealized": 0.5295031070709229,
|
216 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
217 |
"eval_ranking_simple": 0.5367494821548462,
|
218 |
-
"eval_regularize": 1.
|
219 |
-
"eval_runtime":
|
220 |
-
"eval_samples_per_second": 18.
|
221 |
-
"eval_steps_per_second": 1.
|
222 |
-
"eval_wo_beta": 6.
|
223 |
"step": 300
|
224 |
},
|
225 |
{
|
226 |
-
"dpo_loss": 1.
|
227 |
"epoch": 0.9919697685403873,
|
228 |
-
"grad_norm": 81.
|
229 |
"learning_rate": 4.8526047530778175e-06,
|
230 |
-
"logits": -0.
|
231 |
-
"logps": -83.
|
232 |
-
"loss": 1.
|
233 |
-
"objective": 1.
|
234 |
"ranking_idealized": 0.5354166626930237,
|
235 |
"ranking_idealized_expo": 0.5254166722297668,
|
236 |
"ranking_simple": 0.6141666769981384,
|
237 |
-
"regularize": 1.
|
238 |
"step": 350,
|
239 |
-
"wo_beta": 5.
|
240 |
},
|
241 |
{
|
242 |
"epoch": 0.9919697685403873,
|
243 |
-
"eval_dpo_loss": 2.
|
244 |
-
"eval_logits": -1.
|
245 |
-
"eval_logps": -82.
|
246 |
-
"eval_loss": 2.
|
247 |
-
"eval_objective": 2.
|
248 |
"eval_ranking_idealized": 0.5295031070709229,
|
249 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
250 |
-
"eval_ranking_simple": 0.
|
251 |
-
"eval_regularize": 2.
|
252 |
-
"eval_runtime":
|
253 |
-
"eval_samples_per_second":
|
254 |
-
"eval_steps_per_second": 1.
|
255 |
-
"eval_wo_beta": 7.
|
256 |
"step": 350
|
257 |
},
|
258 |
{
|
259 |
-
"dpo_loss": 0.
|
260 |
"epoch": 1.1336797354747283,
|
261 |
-
"grad_norm": 72.
|
262 |
"learning_rate": 4.757316345716554e-06,
|
263 |
-
"logits": -0.
|
264 |
-
"logps": -77.
|
265 |
"loss": 0.725,
|
266 |
-
"objective": 0.
|
267 |
"ranking_idealized": 0.5412499904632568,
|
268 |
"ranking_idealized_expo": 0.5320833325386047,
|
269 |
-
"ranking_simple": 0.
|
270 |
-
"regularize": 0.
|
271 |
"step": 400,
|
272 |
-
"wo_beta": 4.
|
273 |
},
|
274 |
{
|
275 |
"epoch": 1.1336797354747283,
|
276 |
-
"eval_dpo_loss": 2.
|
277 |
-
"eval_logits": -0.
|
278 |
-
"eval_logps": -84.
|
279 |
-
"eval_loss": 2.
|
280 |
-
"eval_objective": 2.
|
281 |
"eval_ranking_idealized": 0.5295031070709229,
|
282 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
283 |
-
"eval_ranking_simple": 0.
|
284 |
-
"eval_regularize": 2.
|
285 |
-
"eval_runtime":
|
286 |
-
"eval_samples_per_second":
|
287 |
-
"eval_steps_per_second": 1.
|
288 |
-
"eval_wo_beta": 7.
|
289 |
"step": 400
|
290 |
},
|
291 |
{
|
292 |
-
"dpo_loss": 0.
|
293 |
"epoch": 1.2753897024090695,
|
294 |
-
"grad_norm": 67.
|
295 |
"learning_rate": 4.639847716126855e-06,
|
296 |
-
"logits": -0.
|
297 |
-
"logps": -
|
298 |
-
"loss": 0.
|
299 |
-
"objective": 0.
|
300 |
"ranking_idealized": 0.5245833396911621,
|
301 |
"ranking_idealized_expo": 0.5191666483879089,
|
302 |
"ranking_simple": 0.6483333110809326,
|
303 |
-
"regularize": 0.
|
304 |
"step": 450,
|
305 |
-
"wo_beta": 5.
|
306 |
},
|
307 |
{
|
308 |
"epoch": 1.2753897024090695,
|
309 |
-
"eval_dpo_loss": 2.
|
310 |
-
"eval_logits": -0.
|
311 |
-
"eval_logps": -80.
|
312 |
-
"eval_loss": 2.
|
313 |
-
"eval_objective": 2.
|
314 |
"eval_ranking_idealized": 0.5295031070709229,
|
315 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
316 |
-
"eval_ranking_simple": 0.
|
317 |
-
"eval_regularize": 2.
|
318 |
-
"eval_runtime":
|
319 |
-
"eval_samples_per_second":
|
320 |
-
"eval_steps_per_second": 1.
|
321 |
-
"eval_wo_beta": 6.
|
322 |
"step": 450
|
323 |
},
|
324 |
{
|
325 |
-
"dpo_loss": 0.
|
326 |
"epoch": 1.4170996693434104,
|
327 |
-
"grad_norm": 80.
|
328 |
"learning_rate": 4.501353102310901e-06,
|
329 |
-
"logits": -0.
|
330 |
-
"logps": -76.
|
331 |
-
"loss": 0.
|
332 |
-
"objective": 0.
|
333 |
"ranking_idealized": 0.5054166913032532,
|
334 |
"ranking_idealized_expo": 0.4970833361148834,
|
335 |
-
"ranking_simple": 0.
|
336 |
-
"regularize": 0.
|
337 |
"step": 500,
|
338 |
-
"wo_beta": 4.
|
339 |
},
|
340 |
{
|
341 |
"epoch": 1.4170996693434104,
|
342 |
-
"eval_dpo_loss": 2.
|
343 |
-
"eval_logits": -0.
|
344 |
-
"eval_logps": -82.
|
345 |
-
"eval_loss": 2.
|
346 |
-
"eval_objective": 2.
|
347 |
"eval_ranking_idealized": 0.5295031070709229,
|
348 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
349 |
-
"eval_ranking_simple": 0.
|
350 |
-
"eval_regularize": 2.
|
351 |
-
"eval_runtime": 308.
|
352 |
-
"eval_samples_per_second": 18.
|
353 |
-
"eval_steps_per_second": 1.
|
354 |
-
"eval_wo_beta": 6.
|
355 |
"step": 500
|
356 |
},
|
357 |
{
|
358 |
-
"dpo_loss": 0.
|
359 |
"epoch": 1.5588096362777515,
|
360 |
-
"grad_norm":
|
361 |
"learning_rate": 4.34319334202531e-06,
|
362 |
-
"logits": -0.
|
363 |
-
"logps": -79.
|
364 |
-
"loss": 0.
|
365 |
-
"objective": 0.
|
366 |
"ranking_idealized": 0.5199999809265137,
|
367 |
"ranking_idealized_expo": 0.5108333230018616,
|
368 |
-
"ranking_simple": 0.
|
369 |
-
"regularize": 0.
|
370 |
"step": 550,
|
371 |
-
"wo_beta": 4.
|
372 |
},
|
373 |
{
|
374 |
"epoch": 1.5588096362777515,
|
375 |
-
"eval_dpo_loss": 2.
|
376 |
-
"eval_logits": -0.
|
377 |
-
"eval_logps": -
|
378 |
-
"eval_loss": 2.
|
379 |
-
"eval_objective": 2.
|
380 |
"eval_ranking_idealized": 0.5295031070709229,
|
381 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
382 |
"eval_ranking_simple": 0.5320910811424255,
|
383 |
-
"eval_regularize": 2.
|
384 |
-
"eval_runtime":
|
385 |
-
"eval_samples_per_second":
|
386 |
-
"eval_steps_per_second": 1.
|
387 |
-
"eval_wo_beta": 6.
|
388 |
"step": 550
|
389 |
},
|
390 |
{
|
391 |
-
"dpo_loss": 0.
|
392 |
"epoch": 1.7005196032120926,
|
393 |
-
"grad_norm": 58.
|
394 |
"learning_rate": 4.16692250129073e-06,
|
395 |
-
"logits": -0.
|
396 |
-
"logps": -79.
|
397 |
-
"loss": 0.
|
398 |
-
"objective": 0.
|
399 |
"ranking_idealized": 0.5220833420753479,
|
400 |
"ranking_idealized_expo": 0.5149999856948853,
|
401 |
-
"ranking_simple": 0.
|
402 |
-
"regularize": 0.
|
403 |
"step": 600,
|
404 |
-
"wo_beta": 4.
|
405 |
},
|
406 |
{
|
407 |
"epoch": 1.7005196032120926,
|
408 |
-
"eval_dpo_loss": 2.
|
409 |
-
"eval_logits": -0.
|
410 |
-
"eval_logps": -81.
|
411 |
-
"eval_loss": 2.
|
412 |
-
"eval_objective": 2.
|
413 |
"eval_ranking_idealized": 0.5295031070709229,
|
414 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
415 |
-
"eval_ranking_simple": 0.
|
416 |
-
"eval_regularize": 2.
|
417 |
-
"eval_runtime": 308.
|
418 |
-
"eval_samples_per_second": 18.
|
419 |
-
"eval_steps_per_second": 1.
|
420 |
-
"eval_wo_beta": 6.
|
421 |
"step": 600
|
422 |
},
|
423 |
{
|
424 |
-
"dpo_loss": 0.
|
425 |
"epoch": 1.8422295701464337,
|
426 |
-
"grad_norm": 71.
|
427 |
"learning_rate": 3.974272604254906e-06,
|
428 |
-
"logits": -0.
|
429 |
-
"logps": -
|
430 |
-
"loss": 0.
|
431 |
-
"objective": 0.
|
432 |
"ranking_idealized": 0.5408333539962769,
|
433 |
"ranking_idealized_expo": 0.527916669845581,
|
434 |
"ranking_simple": 0.6583333611488342,
|
435 |
-
"regularize": 0.
|
436 |
"step": 650,
|
437 |
-
"wo_beta": 4.
|
438 |
},
|
439 |
{
|
440 |
"epoch": 1.8422295701464337,
|
441 |
-
"eval_dpo_loss": 2.
|
442 |
-
"eval_logits": -0.
|
443 |
-
"eval_logps": -82.
|
444 |
-
"eval_loss": 2.
|
445 |
-
"eval_objective": 2.
|
446 |
"eval_ranking_idealized": 0.5295031070709229,
|
447 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
448 |
-
"eval_ranking_simple": 0.
|
449 |
-
"eval_regularize": 2.
|
450 |
-
"eval_runtime":
|
451 |
-
"eval_samples_per_second":
|
452 |
-
"eval_steps_per_second": 1.
|
453 |
-
"eval_wo_beta": 6.
|
454 |
"step": 650
|
455 |
},
|
456 |
{
|
457 |
-
"dpo_loss": 0.
|
458 |
"epoch": 1.9839395370807746,
|
459 |
-
"grad_norm":
|
460 |
"learning_rate": 3.767136614452458e-06,
|
461 |
-
"logits": -0.
|
462 |
-
"logps": -78.
|
463 |
-
"loss": 0.
|
464 |
-
"objective": 0.
|
465 |
"ranking_idealized": 0.5170833468437195,
|
466 |
"ranking_idealized_expo": 0.5112500190734863,
|
467 |
-
"ranking_simple": 0.
|
468 |
-
"regularize": 0.
|
469 |
"step": 700,
|
470 |
-
"wo_beta": 4.
|
471 |
},
|
472 |
{
|
473 |
"epoch": 1.9839395370807746,
|
474 |
-
"eval_dpo_loss": 2.
|
475 |
-
"eval_logits": -0.
|
476 |
-
"eval_logps": -83.
|
477 |
-
"eval_loss": 2.
|
478 |
-
"eval_objective": 2.
|
479 |
"eval_ranking_idealized": 0.5295031070709229,
|
480 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
481 |
-
"eval_ranking_simple": 0.
|
482 |
-
"eval_regularize": 2.
|
483 |
-
"eval_runtime": 308.
|
484 |
-
"eval_samples_per_second": 18.
|
485 |
-
"eval_steps_per_second": 1.
|
486 |
-
"eval_wo_beta": 6.
|
487 |
"step": 700
|
488 |
},
|
489 |
{
|
490 |
-
"dpo_loss": 0.
|
491 |
"epoch": 2.1256495040151155,
|
492 |
-
"grad_norm": 45.
|
493 |
"learning_rate": 3.547549834686222e-06,
|
494 |
-
"logits": -0.
|
495 |
-
"logps": -79.
|
496 |
-
"loss": 0.
|
497 |
-
"objective": 0.
|
498 |
"ranking_idealized": 0.5216666460037231,
|
499 |
"ranking_idealized_expo": 0.5112500190734863,
|
500 |
-
"ranking_simple": 0.
|
501 |
-
"regularize": 0.
|
502 |
"step": 750,
|
503 |
-
"wo_beta": 4.
|
504 |
},
|
505 |
{
|
506 |
"epoch": 2.1256495040151155,
|
507 |
-
"eval_dpo_loss": 2.
|
508 |
-
"eval_logits": -0.
|
509 |
-
"eval_logps": -84.
|
510 |
-
"eval_loss": 2.
|
511 |
-
"eval_objective": 2.
|
512 |
"eval_ranking_idealized": 0.5295031070709229,
|
513 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
514 |
-
"eval_ranking_simple": 0.
|
515 |
-
"eval_regularize": 2.
|
516 |
-
"eval_runtime":
|
517 |
-
"eval_samples_per_second":
|
518 |
-
"eval_steps_per_second": 1.
|
519 |
-
"eval_wo_beta": 6.
|
520 |
"step": 750
|
521 |
},
|
522 |
{
|
523 |
-
"dpo_loss": 0.
|
524 |
"epoch": 2.2673594709494567,
|
525 |
-
"grad_norm": 40.
|
526 |
"learning_rate": 3.3176699082935546e-06,
|
527 |
-
"logits": -0.
|
528 |
-
"logps": -80.
|
529 |
-
"loss": 0.
|
530 |
-
"objective": 0.
|
531 |
"ranking_idealized": 0.5195833444595337,
|
532 |
"ranking_idealized_expo": 0.512499988079071,
|
533 |
-
"ranking_simple": 0.
|
534 |
-
"regularize": 0.
|
535 |
"step": 800,
|
536 |
-
"wo_beta": 3.
|
537 |
},
|
538 |
{
|
539 |
"epoch": 2.2673594709494567,
|
540 |
-
"eval_dpo_loss": 2.
|
541 |
-
"eval_logits": -0.
|
542 |
-
"eval_logps": -84.
|
543 |
-
"eval_loss": 2.
|
544 |
-
"eval_objective": 2.
|
545 |
"eval_ranking_idealized": 0.5295031070709229,
|
546 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
547 |
-
"eval_ranking_simple": 0.
|
548 |
-
"eval_regularize": 2.
|
549 |
-
"eval_runtime":
|
550 |
-
"eval_samples_per_second":
|
551 |
-
"eval_steps_per_second": 1.
|
552 |
-
"eval_wo_beta": 6.
|
553 |
"step": 800
|
554 |
},
|
555 |
{
|
556 |
-
"dpo_loss": 0.
|
557 |
"epoch": 2.409069437883798,
|
558 |
-
"grad_norm":
|
559 |
"learning_rate": 3.0797556183036582e-06,
|
560 |
-
"logits": -0.
|
561 |
-
"logps": -80.
|
562 |
-
"loss": 0.
|
563 |
-
"objective": 0.
|
564 |
"ranking_idealized": 0.5254166722297668,
|
565 |
"ranking_idealized_expo": 0.5141666531562805,
|
566 |
-
"ranking_simple": 0.
|
567 |
-
"regularize": 0.
|
568 |
"step": 850,
|
569 |
-
"wo_beta": 4.
|
570 |
},
|
571 |
{
|
572 |
"epoch": 2.409069437883798,
|
573 |
-
"eval_dpo_loss": 2.
|
574 |
-
"eval_logits": -0.
|
575 |
-
"eval_logps": -
|
576 |
-
"eval_loss": 2.
|
577 |
-
"eval_objective": 2.
|
578 |
"eval_ranking_idealized": 0.5295031070709229,
|
579 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
580 |
-
"eval_ranking_simple": 0.
|
581 |
-
"eval_regularize": 2.
|
582 |
-
"eval_runtime": 308.
|
583 |
-
"eval_samples_per_second": 18.
|
584 |
-
"eval_steps_per_second": 1.
|
585 |
-
"eval_wo_beta": 6.
|
586 |
"step": 850
|
587 |
},
|
588 |
{
|
589 |
-
"dpo_loss": 0.
|
590 |
"epoch": 2.550779404818139,
|
591 |
-
"grad_norm":
|
592 |
"learning_rate": 2.8361446928038298e-06,
|
593 |
-
"logits": -0.
|
594 |
-
"logps": -79.
|
595 |
-
"loss": 0.
|
596 |
-
"objective": 0.
|
597 |
"ranking_idealized": 0.5249999761581421,
|
598 |
"ranking_idealized_expo": 0.5183333158493042,
|
599 |
-
"ranking_simple": 0.
|
600 |
-
"regularize": 0.
|
601 |
"step": 900,
|
602 |
-
"wo_beta": 4.
|
603 |
},
|
604 |
{
|
605 |
"epoch": 2.550779404818139,
|
606 |
-
"eval_dpo_loss": 2.
|
607 |
-
"eval_logits": -0.
|
608 |
-
"eval_logps": -84.
|
609 |
-
"eval_loss": 2.
|
610 |
-
"eval_objective": 2.
|
611 |
"eval_ranking_idealized": 0.5295031070709229,
|
612 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
613 |
-
"eval_ranking_simple": 0.
|
614 |
-
"eval_regularize": 2.
|
615 |
-
"eval_runtime":
|
616 |
-
"eval_samples_per_second":
|
617 |
-
"eval_steps_per_second": 1.
|
618 |
-
"eval_wo_beta": 6.
|
619 |
"step": 900
|
620 |
},
|
621 |
{
|
622 |
-
"dpo_loss": 0.
|
623 |
"epoch": 2.69248937175248,
|
624 |
-
"grad_norm":
|
625 |
"learning_rate": 2.5892308345974517e-06,
|
626 |
-
"logits": -0.
|
627 |
-
"logps": -
|
628 |
-
"loss": 0.
|
629 |
-
"objective": 0.
|
630 |
"ranking_idealized": 0.5162500143051147,
|
631 |
"ranking_idealized_expo": 0.5058333277702332,
|
632 |
-
"ranking_simple": 0.
|
633 |
-
"regularize": 0.
|
634 |
"step": 950,
|
635 |
-
"wo_beta": 4.
|
636 |
},
|
637 |
{
|
638 |
"epoch": 2.69248937175248,
|
639 |
-
"eval_dpo_loss": 2.
|
640 |
-
"eval_logits": -0.
|
641 |
-
"eval_logps": -83.
|
642 |
-
"eval_loss": 2.
|
643 |
-
"eval_objective": 2.
|
644 |
"eval_ranking_idealized": 0.5295031070709229,
|
645 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
646 |
-
"eval_ranking_simple": 0.
|
647 |
-
"eval_regularize": 2.
|
648 |
-
"eval_runtime":
|
649 |
-
"eval_samples_per_second": 18.
|
650 |
-
"eval_steps_per_second": 1.
|
651 |
-
"eval_wo_beta": 6.
|
652 |
"step": 950
|
653 |
},
|
654 |
{
|
655 |
"epoch": 2.69248937175248,
|
656 |
"step": 950,
|
657 |
"total_flos": 0.0,
|
658 |
-
"train_loss": 0.
|
659 |
-
"train_runtime":
|
660 |
-
"train_samples_per_second": 9.
|
661 |
-
"train_steps_per_second": 0.
|
662 |
}
|
663 |
],
|
664 |
"logging_steps": 50,
|
|
|
1 |
{
|
2 |
+
"best_metric": 6.68360710144043,
|
3 |
"best_model_checkpoint": "./qwen2.5-0.5b/qwen2.5-0.5b-expo-DPO-ES-1/checkpoint-700",
|
4 |
"epoch": 2.69248937175248,
|
5 |
"eval_steps": 50,
|
|
|
11 |
{
|
12 |
"dpo_loss": 0.6931471824645996,
|
13 |
"epoch": 0.002834199338686821,
|
14 |
+
"grad_norm": 184.42504161174296,
|
15 |
"learning_rate": 2.840909090909091e-08,
|
16 |
"logits": -1.359458565711975,
|
17 |
"logps": -84.69721221923828,
|
|
|
25 |
"wo_beta": 5.271125316619873
|
26 |
},
|
27 |
{
|
28 |
+
"dpo_loss": 0.7105370759963989,
|
29 |
"epoch": 0.14170996693434104,
|
30 |
+
"grad_norm": 184.8690438155141,
|
31 |
"learning_rate": 1.4204545454545458e-06,
|
32 |
+
"logits": -1.464050531387329,
|
33 |
+
"logps": -84.1650161743164,
|
34 |
"loss": 0.7017,
|
35 |
+
"objective": 0.7105370759963989,
|
36 |
"ranking_idealized": 0.5289115905761719,
|
37 |
"ranking_idealized_expo": 0.5221088528633118,
|
38 |
"ranking_simple": 0.5225340127944946,
|
39 |
+
"regularize": 0.7105370759963989,
|
40 |
"step": 50,
|
41 |
+
"wo_beta": 7.069356918334961
|
42 |
},
|
43 |
{
|
44 |
"epoch": 0.14170996693434104,
|
45 |
+
"eval_dpo_loss": 0.856997013092041,
|
46 |
+
"eval_logits": -1.458188533782959,
|
47 |
+
"eval_logps": -93.02446746826172,
|
48 |
+
"eval_loss": 0.8469988107681274,
|
49 |
+
"eval_objective": 0.856997013092041,
|
50 |
"eval_ranking_idealized": 0.5295031070709229,
|
51 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
52 |
"eval_ranking_simple": 0.523809552192688,
|
53 |
+
"eval_regularize": 0.856997013092041,
|
54 |
+
"eval_runtime": 312.0194,
|
55 |
+
"eval_samples_per_second": 18.557,
|
56 |
+
"eval_steps_per_second": 1.548,
|
57 |
+
"eval_wo_beta": 7.850645065307617,
|
58 |
"step": 50
|
59 |
},
|
60 |
{
|
61 |
+
"dpo_loss": 0.8048794865608215,
|
62 |
"epoch": 0.2834199338686821,
|
63 |
+
"grad_norm": 165.00318532005144,
|
64 |
"learning_rate": 2.8409090909090916e-06,
|
65 |
+
"logits": -1.4549062252044678,
|
66 |
+
"logps": -84.21845245361328,
|
67 |
"loss": 0.8112,
|
68 |
+
"objective": 0.8048794865608215,
|
69 |
"ranking_idealized": 0.5241666436195374,
|
70 |
"ranking_idealized_expo": 0.5137500166893005,
|
71 |
"ranking_simple": 0.543749988079071,
|
72 |
+
"regularize": 0.8048794865608215,
|
73 |
"step": 100,
|
74 |
"wo_beta": 6.3764801025390625
|
75 |
},
|
76 |
{
|
77 |
"epoch": 0.2834199338686821,
|
78 |
+
"eval_dpo_loss": 1.0273518562316895,
|
79 |
+
"eval_logits": -1.4382514953613281,
|
80 |
+
"eval_logps": -86.68037414550781,
|
81 |
+
"eval_loss": 1.0529025793075562,
|
82 |
+
"eval_objective": 1.0273518562316895,
|
83 |
"eval_ranking_idealized": 0.5295031070709229,
|
84 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
85 |
"eval_ranking_simple": 0.5284678936004639,
|
86 |
+
"eval_regularize": 1.0273518562316895,
|
87 |
+
"eval_runtime": 308.916,
|
88 |
+
"eval_samples_per_second": 18.743,
|
89 |
+
"eval_steps_per_second": 1.564,
|
90 |
+
"eval_wo_beta": 7.498167991638184,
|
91 |
"step": 100
|
92 |
},
|
93 |
{
|
94 |
+
"dpo_loss": 1.0717324018478394,
|
95 |
"epoch": 0.42512990080302315,
|
96 |
+
"grad_norm": 132.13237608472397,
|
97 |
"learning_rate": 4.2613636363636365e-06,
|
98 |
+
"logits": -1.3764770030975342,
|
99 |
+
"logps": -75.389892578125,
|
100 |
"loss": 1.0895,
|
101 |
+
"objective": 1.0717324018478394,
|
102 |
"ranking_idealized": 0.5333333611488342,
|
103 |
"ranking_idealized_expo": 0.527916669845581,
|
104 |
"ranking_simple": 0.5645833611488342,
|
105 |
+
"regularize": 1.0717324018478394,
|
106 |
"step": 150,
|
107 |
+
"wo_beta": 6.200186729431152
|
108 |
},
|
109 |
{
|
110 |
"epoch": 0.42512990080302315,
|
111 |
+
"eval_dpo_loss": 1.4009839296340942,
|
112 |
+
"eval_logits": -1.2964812517166138,
|
113 |
+
"eval_logps": -84.43347930908203,
|
114 |
+
"eval_loss": 1.4497517347335815,
|
115 |
+
"eval_objective": 1.4009839296340942,
|
116 |
"eval_ranking_idealized": 0.5295031070709229,
|
117 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
118 |
"eval_ranking_simple": 0.5320910811424255,
|
119 |
+
"eval_regularize": 1.4009839296340942,
|
120 |
+
"eval_runtime": 308.861,
|
121 |
+
"eval_samples_per_second": 18.746,
|
122 |
+
"eval_steps_per_second": 1.564,
|
123 |
+
"eval_wo_beta": 7.269246578216553,
|
124 |
"step": 150
|
125 |
},
|
126 |
{
|
127 |
+
"dpo_loss": 1.2378751039505005,
|
128 |
"epoch": 0.5668398677373642,
|
129 |
+
"grad_norm": 120.88819473175299,
|
130 |
"learning_rate": 4.997168347957521e-06,
|
131 |
+
"logits": -1.3378355503082275,
|
132 |
+
"logps": -75.85352325439453,
|
133 |
+
"loss": 1.2362,
|
134 |
+
"objective": 1.2378751039505005,
|
135 |
"ranking_idealized": 0.5204166769981384,
|
136 |
"ranking_idealized_expo": 0.51541668176651,
|
137 |
"ranking_simple": 0.5699999928474426,
|
138 |
+
"regularize": 1.2378751039505005,
|
139 |
"step": 200,
|
140 |
+
"wo_beta": 6.1314239501953125
|
141 |
},
|
142 |
{
|
143 |
"epoch": 0.5668398677373642,
|
144 |
+
"eval_dpo_loss": 1.611555576324463,
|
145 |
+
"eval_logits": -1.2954967021942139,
|
146 |
+
"eval_logps": -77.71936798095703,
|
147 |
+
"eval_loss": 1.7034811973571777,
|
148 |
+
"eval_objective": 1.611555576324463,
|
149 |
"eval_ranking_idealized": 0.5295031070709229,
|
150 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
151 |
"eval_ranking_simple": 0.5320910811424255,
|
152 |
+
"eval_regularize": 1.611555576324463,
|
153 |
+
"eval_runtime": 309.5132,
|
154 |
+
"eval_samples_per_second": 18.707,
|
155 |
+
"eval_steps_per_second": 1.561,
|
156 |
+
"eval_wo_beta": 7.226413726806641,
|
157 |
"step": 200
|
158 |
},
|
159 |
{
|
160 |
+
"dpo_loss": 1.2734363079071045,
|
161 |
"epoch": 0.7085498346717053,
|
162 |
+
"grad_norm": 126.97448798013814,
|
163 |
"learning_rate": 4.973122855144066e-06,
|
164 |
+
"logits": -1.1745147705078125,
|
165 |
+
"logps": -80.53407287597656,
|
166 |
+
"loss": 1.3151,
|
167 |
+
"objective": 1.2734363079071045,
|
168 |
"ranking_idealized": 0.5249999761581421,
|
169 |
"ranking_idealized_expo": 0.5162500143051147,
|
170 |
"ranking_simple": 0.5929166674613953,
|
171 |
+
"regularize": 1.2734363079071045,
|
172 |
"step": 250,
|
173 |
+
"wo_beta": 6.07757568359375
|
174 |
},
|
175 |
{
|
176 |
"epoch": 0.7085498346717053,
|
177 |
+
"eval_dpo_loss": 1.831916093826294,
|
178 |
+
"eval_logits": -1.2565279006958008,
|
179 |
+
"eval_logps": -92.72237396240234,
|
180 |
+
"eval_loss": 1.9222025871276855,
|
181 |
+
"eval_objective": 1.831916093826294,
|
182 |
"eval_ranking_idealized": 0.5295031070709229,
|
183 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
184 |
"eval_ranking_simple": 0.5310559272766113,
|
185 |
+
"eval_regularize": 1.831916093826294,
|
186 |
+
"eval_runtime": 308.4533,
|
187 |
+
"eval_samples_per_second": 18.771,
|
188 |
+
"eval_steps_per_second": 1.566,
|
189 |
+
"eval_wo_beta": 7.185470104217529,
|
190 |
"step": 250
|
191 |
},
|
192 |
{
|
193 |
+
"dpo_loss": 1.1058416366577148,
|
194 |
"epoch": 0.8502598016060463,
|
195 |
+
"grad_norm": 93.62402691602496,
|
196 |
"learning_rate": 4.924776641419513e-06,
|
197 |
+
"logits": -1.027516484260559,
|
198 |
+
"logps": -86.21515655517578,
|
199 |
"loss": 1.1899,
|
200 |
+
"objective": 1.1058416366577148,
|
201 |
"ranking_idealized": 0.5062500238418579,
|
202 |
"ranking_idealized_expo": 0.4950000047683716,
|
203 |
"ranking_simple": 0.5833333134651184,
|
204 |
+
"regularize": 1.1058416366577148,
|
205 |
"step": 300,
|
206 |
+
"wo_beta": 5.9321818351745605
|
207 |
},
|
208 |
{
|
209 |
"epoch": 0.8502598016060463,
|
210 |
+
"eval_dpo_loss": 1.9587332010269165,
|
211 |
+
"eval_logits": -0.9785559177398682,
|
212 |
+
"eval_logps": -90.93506622314453,
|
213 |
+
"eval_loss": 2.0297279357910156,
|
214 |
+
"eval_objective": 1.9587332010269165,
|
215 |
"eval_ranking_idealized": 0.5295031070709229,
|
216 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
217 |
"eval_ranking_simple": 0.5367494821548462,
|
218 |
+
"eval_regularize": 1.9587332010269165,
|
219 |
+
"eval_runtime": 308.5179,
|
220 |
+
"eval_samples_per_second": 18.767,
|
221 |
+
"eval_steps_per_second": 1.566,
|
222 |
+
"eval_wo_beta": 6.933654308319092,
|
223 |
"step": 300
|
224 |
},
|
225 |
{
|
226 |
+
"dpo_loss": 1.1039565801620483,
|
227 |
"epoch": 0.9919697685403873,
|
228 |
+
"grad_norm": 81.61244111449733,
|
229 |
"learning_rate": 4.8526047530778175e-06,
|
230 |
+
"logits": -0.8597685694694519,
|
231 |
+
"logps": -83.09705352783203,
|
232 |
+
"loss": 1.1441,
|
233 |
+
"objective": 1.1039565801620483,
|
234 |
"ranking_idealized": 0.5354166626930237,
|
235 |
"ranking_idealized_expo": 0.5254166722297668,
|
236 |
"ranking_simple": 0.6141666769981384,
|
237 |
+
"regularize": 1.1039565801620483,
|
238 |
"step": 350,
|
239 |
+
"wo_beta": 5.234945297241211
|
240 |
},
|
241 |
{
|
242 |
"epoch": 0.9919697685403873,
|
243 |
+
"eval_dpo_loss": 2.054466485977173,
|
244 |
+
"eval_logits": -1.0211108922958374,
|
245 |
+
"eval_logps": -82.12905883789062,
|
246 |
+
"eval_loss": 2.1653144359588623,
|
247 |
+
"eval_objective": 2.054466485977173,
|
248 |
"eval_ranking_idealized": 0.5295031070709229,
|
249 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
250 |
+
"eval_ranking_simple": 0.542443037033081,
|
251 |
+
"eval_regularize": 2.054466485977173,
|
252 |
+
"eval_runtime": 308.5833,
|
253 |
+
"eval_samples_per_second": 18.763,
|
254 |
+
"eval_steps_per_second": 1.565,
|
255 |
+
"eval_wo_beta": 7.001650810241699,
|
256 |
"step": 350
|
257 |
},
|
258 |
{
|
259 |
+
"dpo_loss": 0.7012167572975159,
|
260 |
"epoch": 1.1336797354747283,
|
261 |
+
"grad_norm": 72.06842141832568,
|
262 |
"learning_rate": 4.757316345716554e-06,
|
263 |
+
"logits": -0.8224175572395325,
|
264 |
+
"logps": -77.92687225341797,
|
265 |
"loss": 0.725,
|
266 |
+
"objective": 0.7012167572975159,
|
267 |
"ranking_idealized": 0.5412499904632568,
|
268 |
"ranking_idealized_expo": 0.5320833325386047,
|
269 |
+
"ranking_simple": 0.6483333110809326,
|
270 |
+
"regularize": 0.7012167572975159,
|
271 |
"step": 400,
|
272 |
+
"wo_beta": 4.492737770080566
|
273 |
},
|
274 |
{
|
275 |
"epoch": 1.1336797354747283,
|
276 |
+
"eval_dpo_loss": 2.2359888553619385,
|
277 |
+
"eval_logits": -0.7528972029685974,
|
278 |
+
"eval_logps": -84.34579467773438,
|
279 |
+
"eval_loss": 2.288560152053833,
|
280 |
+
"eval_objective": 2.2359888553619385,
|
281 |
"eval_ranking_idealized": 0.5295031070709229,
|
282 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
283 |
+
"eval_ranking_simple": 0.5331262946128845,
|
284 |
+
"eval_regularize": 2.2359888553619385,
|
285 |
+
"eval_runtime": 308.7822,
|
286 |
+
"eval_samples_per_second": 18.751,
|
287 |
+
"eval_steps_per_second": 1.564,
|
288 |
+
"eval_wo_beta": 7.154054164886475,
|
289 |
"step": 400
|
290 |
},
|
291 |
{
|
292 |
+
"dpo_loss": 0.78798508644104,
|
293 |
"epoch": 1.2753897024090695,
|
294 |
+
"grad_norm": 67.40926541067125,
|
295 |
"learning_rate": 4.639847716126855e-06,
|
296 |
+
"logits": -0.7312514185905457,
|
297 |
+
"logps": -80.1772232055664,
|
298 |
+
"loss": 0.7626,
|
299 |
+
"objective": 0.78798508644104,
|
300 |
"ranking_idealized": 0.5245833396911621,
|
301 |
"ranking_idealized_expo": 0.5191666483879089,
|
302 |
"ranking_simple": 0.6483333110809326,
|
303 |
+
"regularize": 0.78798508644104,
|
304 |
"step": 450,
|
305 |
+
"wo_beta": 5.0193963050842285
|
306 |
},
|
307 |
{
|
308 |
"epoch": 1.2753897024090695,
|
309 |
+
"eval_dpo_loss": 2.0657246112823486,
|
310 |
+
"eval_logits": -0.8863243460655212,
|
311 |
+
"eval_logps": -80.59545135498047,
|
312 |
+
"eval_loss": 2.1594510078430176,
|
313 |
+
"eval_objective": 2.0657246112823486,
|
314 |
"eval_ranking_idealized": 0.5295031070709229,
|
315 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
316 |
+
"eval_ranking_simple": 0.532608687877655,
|
317 |
+
"eval_regularize": 2.0657246112823486,
|
318 |
+
"eval_runtime": 314.8953,
|
319 |
+
"eval_samples_per_second": 18.387,
|
320 |
+
"eval_steps_per_second": 1.534,
|
321 |
+
"eval_wo_beta": 6.7939372062683105,
|
322 |
"step": 450
|
323 |
},
|
324 |
{
|
325 |
+
"dpo_loss": 0.7955495119094849,
|
326 |
"epoch": 1.4170996693434104,
|
327 |
+
"grad_norm": 80.21112193307313,
|
328 |
"learning_rate": 4.501353102310901e-06,
|
329 |
+
"logits": -0.717734694480896,
|
330 |
+
"logps": -76.27667999267578,
|
331 |
+
"loss": 0.8048,
|
332 |
+
"objective": 0.7955495119094849,
|
333 |
"ranking_idealized": 0.5054166913032532,
|
334 |
"ranking_idealized_expo": 0.4970833361148834,
|
335 |
+
"ranking_simple": 0.6424999833106995,
|
336 |
+
"regularize": 0.7955495119094849,
|
337 |
"step": 500,
|
338 |
+
"wo_beta": 4.868187427520752
|
339 |
},
|
340 |
{
|
341 |
"epoch": 1.4170996693434104,
|
342 |
+
"eval_dpo_loss": 2.0975162982940674,
|
343 |
+
"eval_logits": -0.7432325482368469,
|
344 |
+
"eval_logps": -82.348876953125,
|
345 |
+
"eval_loss": 2.2134060859680176,
|
346 |
+
"eval_objective": 2.0975162982940674,
|
347 |
"eval_ranking_idealized": 0.5295031070709229,
|
348 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
349 |
+
"eval_ranking_simple": 0.5341615080833435,
|
350 |
+
"eval_regularize": 2.0975162982940674,
|
351 |
+
"eval_runtime": 308.5741,
|
352 |
+
"eval_samples_per_second": 18.764,
|
353 |
+
"eval_steps_per_second": 1.565,
|
354 |
+
"eval_wo_beta": 6.79838752746582,
|
355 |
"step": 500
|
356 |
},
|
357 |
{
|
358 |
+
"dpo_loss": 0.733691394329071,
|
359 |
"epoch": 1.5588096362777515,
|
360 |
+
"grad_norm": 73.44558837421432,
|
361 |
"learning_rate": 4.34319334202531e-06,
|
362 |
+
"logits": -0.6085548996925354,
|
363 |
+
"logps": -79.48100280761719,
|
364 |
+
"loss": 0.7106,
|
365 |
+
"objective": 0.733691394329071,
|
366 |
"ranking_idealized": 0.5199999809265137,
|
367 |
"ranking_idealized_expo": 0.5108333230018616,
|
368 |
+
"ranking_simple": 0.6579166650772095,
|
369 |
+
"regularize": 0.733691394329071,
|
370 |
"step": 550,
|
371 |
+
"wo_beta": 4.594529151916504
|
372 |
},
|
373 |
{
|
374 |
"epoch": 1.5588096362777515,
|
375 |
+
"eval_dpo_loss": 2.069565534591675,
|
376 |
+
"eval_logits": -0.6665427088737488,
|
377 |
+
"eval_logps": -85.06731414794922,
|
378 |
+
"eval_loss": 2.1704676151275635,
|
379 |
+
"eval_objective": 2.069565534591675,
|
380 |
"eval_ranking_idealized": 0.5295031070709229,
|
381 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
382 |
"eval_ranking_simple": 0.5320910811424255,
|
383 |
+
"eval_regularize": 2.069565534591675,
|
384 |
+
"eval_runtime": 308.5,
|
385 |
+
"eval_samples_per_second": 18.768,
|
386 |
+
"eval_steps_per_second": 1.566,
|
387 |
+
"eval_wo_beta": 6.861428260803223,
|
388 |
"step": 550
|
389 |
},
|
390 |
{
|
391 |
+
"dpo_loss": 0.6737102270126343,
|
392 |
"epoch": 1.7005196032120926,
|
393 |
+
"grad_norm": 58.923373544475695,
|
394 |
"learning_rate": 4.16692250129073e-06,
|
395 |
+
"logits": -0.5712339282035828,
|
396 |
+
"logps": -79.50873565673828,
|
397 |
+
"loss": 0.6934,
|
398 |
+
"objective": 0.6737102270126343,
|
399 |
"ranking_idealized": 0.5220833420753479,
|
400 |
"ranking_idealized_expo": 0.5149999856948853,
|
401 |
+
"ranking_simple": 0.6537500023841858,
|
402 |
+
"regularize": 0.6737102270126343,
|
403 |
"step": 600,
|
404 |
+
"wo_beta": 4.632404804229736
|
405 |
},
|
406 |
{
|
407 |
"epoch": 1.7005196032120926,
|
408 |
+
"eval_dpo_loss": 2.0693418979644775,
|
409 |
+
"eval_logits": -0.7358241081237793,
|
410 |
+
"eval_logps": -81.67726135253906,
|
411 |
+
"eval_loss": 2.2126805782318115,
|
412 |
+
"eval_objective": 2.0693418979644775,
|
413 |
"eval_ranking_idealized": 0.5295031070709229,
|
414 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
415 |
+
"eval_ranking_simple": 0.5362318754196167,
|
416 |
+
"eval_regularize": 2.0693418979644775,
|
417 |
+
"eval_runtime": 308.5887,
|
418 |
+
"eval_samples_per_second": 18.763,
|
419 |
+
"eval_steps_per_second": 1.565,
|
420 |
+
"eval_wo_beta": 6.726527690887451,
|
421 |
"step": 600
|
422 |
},
|
423 |
{
|
424 |
+
"dpo_loss": 0.6718880534172058,
|
425 |
"epoch": 1.8422295701464337,
|
426 |
+
"grad_norm": 71.22185869553977,
|
427 |
"learning_rate": 3.974272604254906e-06,
|
428 |
+
"logits": -0.6514729261398315,
|
429 |
+
"logps": -79.11984252929688,
|
430 |
+
"loss": 0.6885,
|
431 |
+
"objective": 0.6718880534172058,
|
432 |
"ranking_idealized": 0.5408333539962769,
|
433 |
"ranking_idealized_expo": 0.527916669845581,
|
434 |
"ranking_simple": 0.6583333611488342,
|
435 |
+
"regularize": 0.6718880534172058,
|
436 |
"step": 650,
|
437 |
+
"wo_beta": 4.779858112335205
|
438 |
},
|
439 |
{
|
440 |
"epoch": 1.8422295701464337,
|
441 |
+
"eval_dpo_loss": 2.1432294845581055,
|
442 |
+
"eval_logits": -0.6786962747573853,
|
443 |
+
"eval_logps": -82.88699340820312,
|
444 |
+
"eval_loss": 2.2198147773742676,
|
445 |
+
"eval_objective": 2.1432294845581055,
|
446 |
"eval_ranking_idealized": 0.5295031070709229,
|
447 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
448 |
+
"eval_ranking_simple": 0.5362318754196167,
|
449 |
+
"eval_regularize": 2.1432294845581055,
|
450 |
+
"eval_runtime": 308.8348,
|
451 |
+
"eval_samples_per_second": 18.748,
|
452 |
+
"eval_steps_per_second": 1.564,
|
453 |
+
"eval_wo_beta": 6.820231914520264,
|
454 |
"step": 650
|
455 |
},
|
456 |
{
|
457 |
+
"dpo_loss": 0.6824830770492554,
|
458 |
"epoch": 1.9839395370807746,
|
459 |
+
"grad_norm": 69.2974675476568,
|
460 |
"learning_rate": 3.767136614452458e-06,
|
461 |
+
"logits": -0.5030444860458374,
|
462 |
+
"logps": -78.93128967285156,
|
463 |
+
"loss": 0.6477,
|
464 |
+
"objective": 0.6824830770492554,
|
465 |
"ranking_idealized": 0.5170833468437195,
|
466 |
"ranking_idealized_expo": 0.5112500190734863,
|
467 |
+
"ranking_simple": 0.6433333158493042,
|
468 |
+
"regularize": 0.6824830770492554,
|
469 |
"step": 700,
|
470 |
+
"wo_beta": 4.807421684265137
|
471 |
},
|
472 |
{
|
473 |
"epoch": 1.9839395370807746,
|
474 |
+
"eval_dpo_loss": 2.2591824531555176,
|
475 |
+
"eval_logits": -0.659694254398346,
|
476 |
+
"eval_logps": -83.41045379638672,
|
477 |
+
"eval_loss": 2.3420462608337402,
|
478 |
+
"eval_objective": 2.2591824531555176,
|
479 |
"eval_ranking_idealized": 0.5295031070709229,
|
480 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
481 |
+
"eval_ranking_simple": 0.5403726696968079,
|
482 |
+
"eval_regularize": 2.2591824531555176,
|
483 |
+
"eval_runtime": 308.6048,
|
484 |
+
"eval_samples_per_second": 18.762,
|
485 |
+
"eval_steps_per_second": 1.565,
|
486 |
+
"eval_wo_beta": 6.68360710144043,
|
487 |
"step": 700
|
488 |
},
|
489 |
{
|
490 |
+
"dpo_loss": 0.38009390234947205,
|
491 |
"epoch": 2.1256495040151155,
|
492 |
+
"grad_norm": 45.305368238738886,
|
493 |
"learning_rate": 3.547549834686222e-06,
|
494 |
+
"logits": -0.5769148468971252,
|
495 |
+
"logps": -79.54972076416016,
|
496 |
+
"loss": 0.3785,
|
497 |
+
"objective": 0.38009390234947205,
|
498 |
"ranking_idealized": 0.5216666460037231,
|
499 |
"ranking_idealized_expo": 0.5112500190734863,
|
500 |
+
"ranking_simple": 0.6787499785423279,
|
501 |
+
"regularize": 0.38009390234947205,
|
502 |
"step": 750,
|
503 |
+
"wo_beta": 4.319232940673828
|
504 |
},
|
505 |
{
|
506 |
"epoch": 2.1256495040151155,
|
507 |
+
"eval_dpo_loss": 2.2005364894866943,
|
508 |
+
"eval_logits": -0.7841140627861023,
|
509 |
+
"eval_logps": -84.03689575195312,
|
510 |
+
"eval_loss": 2.2919445037841797,
|
511 |
+
"eval_objective": 2.2005364894866943,
|
512 |
"eval_ranking_idealized": 0.5295031070709229,
|
513 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
514 |
+
"eval_ranking_simple": 0.54347825050354,
|
515 |
+
"eval_regularize": 2.2005364894866943,
|
516 |
+
"eval_runtime": 311.4559,
|
517 |
+
"eval_samples_per_second": 18.59,
|
518 |
+
"eval_steps_per_second": 1.551,
|
519 |
+
"eval_wo_beta": 6.8514323234558105,
|
520 |
"step": 750
|
521 |
},
|
522 |
{
|
523 |
+
"dpo_loss": 0.33919599652290344,
|
524 |
"epoch": 2.2673594709494567,
|
525 |
+
"grad_norm": 40.098481963513144,
|
526 |
"learning_rate": 3.3176699082935546e-06,
|
527 |
+
"logits": -0.6165332198143005,
|
528 |
+
"logps": -80.33509826660156,
|
529 |
+
"loss": 0.3316,
|
530 |
+
"objective": 0.33919599652290344,
|
531 |
"ranking_idealized": 0.5195833444595337,
|
532 |
"ranking_idealized_expo": 0.512499988079071,
|
533 |
+
"ranking_simple": 0.6795833110809326,
|
534 |
+
"regularize": 0.33919599652290344,
|
535 |
"step": 800,
|
536 |
+
"wo_beta": 3.9735937118530273
|
537 |
},
|
538 |
{
|
539 |
"epoch": 2.2673594709494567,
|
540 |
+
"eval_dpo_loss": 2.1123316287994385,
|
541 |
+
"eval_logits": -0.6766893267631531,
|
542 |
+
"eval_logps": -84.29901123046875,
|
543 |
+
"eval_loss": 2.2220041751861572,
|
544 |
+
"eval_objective": 2.1123316287994385,
|
545 |
"eval_ranking_idealized": 0.5295031070709229,
|
546 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
547 |
+
"eval_ranking_simple": 0.5408902764320374,
|
548 |
+
"eval_regularize": 2.1123316287994385,
|
549 |
+
"eval_runtime": 309.0664,
|
550 |
+
"eval_samples_per_second": 18.734,
|
551 |
+
"eval_steps_per_second": 1.563,
|
552 |
+
"eval_wo_beta": 6.766322612762451,
|
553 |
"step": 800
|
554 |
},
|
555 |
{
|
556 |
+
"dpo_loss": 0.34950682520866394,
|
557 |
"epoch": 2.409069437883798,
|
558 |
+
"grad_norm": 35.45933765377846,
|
559 |
"learning_rate": 3.0797556183036582e-06,
|
560 |
+
"logits": -0.6031672954559326,
|
561 |
+
"logps": -80.82646942138672,
|
562 |
+
"loss": 0.3283,
|
563 |
+
"objective": 0.34950682520866394,
|
564 |
"ranking_idealized": 0.5254166722297668,
|
565 |
"ranking_idealized_expo": 0.5141666531562805,
|
566 |
+
"ranking_simple": 0.6779166460037231,
|
567 |
+
"regularize": 0.34950682520866394,
|
568 |
"step": 850,
|
569 |
+
"wo_beta": 4.280696392059326
|
570 |
},
|
571 |
{
|
572 |
"epoch": 2.409069437883798,
|
573 |
+
"eval_dpo_loss": 2.221233606338501,
|
574 |
+
"eval_logits": -0.6537899374961853,
|
575 |
+
"eval_logps": -85.08342742919922,
|
576 |
+
"eval_loss": 2.3019895553588867,
|
577 |
+
"eval_objective": 2.221233606338501,
|
578 |
"eval_ranking_idealized": 0.5295031070709229,
|
579 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
580 |
+
"eval_ranking_simple": 0.5408902764320374,
|
581 |
+
"eval_regularize": 2.221233606338501,
|
582 |
+
"eval_runtime": 308.3167,
|
583 |
+
"eval_samples_per_second": 18.779,
|
584 |
+
"eval_steps_per_second": 1.567,
|
585 |
+
"eval_wo_beta": 6.777287006378174,
|
586 |
"step": 850
|
587 |
},
|
588 |
{
|
589 |
+
"dpo_loss": 0.35370326042175293,
|
590 |
"epoch": 2.550779404818139,
|
591 |
+
"grad_norm": 63.704496862923826,
|
592 |
"learning_rate": 2.8361446928038298e-06,
|
593 |
+
"logits": -0.600394606590271,
|
594 |
+
"logps": -79.94818115234375,
|
595 |
+
"loss": 0.3516,
|
596 |
+
"objective": 0.35370326042175293,
|
597 |
"ranking_idealized": 0.5249999761581421,
|
598 |
"ranking_idealized_expo": 0.5183333158493042,
|
599 |
+
"ranking_simple": 0.6729166507720947,
|
600 |
+
"regularize": 0.35370326042175293,
|
601 |
"step": 900,
|
602 |
+
"wo_beta": 4.2681732177734375
|
603 |
},
|
604 |
{
|
605 |
"epoch": 2.550779404818139,
|
606 |
+
"eval_dpo_loss": 2.1911425590515137,
|
607 |
+
"eval_logits": -0.6225402355194092,
|
608 |
+
"eval_logps": -84.7564468383789,
|
609 |
+
"eval_loss": 2.272311210632324,
|
610 |
+
"eval_objective": 2.1911425590515137,
|
611 |
"eval_ranking_idealized": 0.5295031070709229,
|
612 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
613 |
+
"eval_ranking_simple": 0.5362318754196167,
|
614 |
+
"eval_regularize": 2.1911425590515137,
|
615 |
+
"eval_runtime": 308.2356,
|
616 |
+
"eval_samples_per_second": 18.784,
|
617 |
+
"eval_steps_per_second": 1.567,
|
618 |
+
"eval_wo_beta": 6.816178798675537,
|
619 |
"step": 900
|
620 |
},
|
621 |
{
|
622 |
+
"dpo_loss": 0.38168859481811523,
|
623 |
"epoch": 2.69248937175248,
|
624 |
+
"grad_norm": 42.960212326170364,
|
625 |
"learning_rate": 2.5892308345974517e-06,
|
626 |
+
"logits": -0.5876100659370422,
|
627 |
+
"logps": -80.23540496826172,
|
628 |
+
"loss": 0.3245,
|
629 |
+
"objective": 0.38168859481811523,
|
630 |
"ranking_idealized": 0.5162500143051147,
|
631 |
"ranking_idealized_expo": 0.5058333277702332,
|
632 |
+
"ranking_simple": 0.6866666674613953,
|
633 |
+
"regularize": 0.38168859481811523,
|
634 |
"step": 950,
|
635 |
+
"wo_beta": 4.207254409790039
|
636 |
},
|
637 |
{
|
638 |
"epoch": 2.69248937175248,
|
639 |
+
"eval_dpo_loss": 2.2523224353790283,
|
640 |
+
"eval_logits": -0.7128884196281433,
|
641 |
+
"eval_logps": -83.6421127319336,
|
642 |
+
"eval_loss": 2.3303723335266113,
|
643 |
+
"eval_objective": 2.2523224353790283,
|
644 |
"eval_ranking_idealized": 0.5295031070709229,
|
645 |
"eval_ranking_idealized_expo": 0.5212215185165405,
|
646 |
+
"eval_ranking_simple": 0.533643901348114,
|
647 |
+
"eval_regularize": 2.2523224353790283,
|
648 |
+
"eval_runtime": 308.3878,
|
649 |
+
"eval_samples_per_second": 18.775,
|
650 |
+
"eval_steps_per_second": 1.566,
|
651 |
+
"eval_wo_beta": 6.89421272277832,
|
652 |
"step": 950
|
653 |
},
|
654 |
{
|
655 |
"epoch": 2.69248937175248,
|
656 |
"step": 950,
|
657 |
"total_flos": 0.0,
|
658 |
+
"train_loss": 0.7491915085441188,
|
659 |
+
"train_runtime": 26152.6935,
|
660 |
+
"train_samples_per_second": 9.713,
|
661 |
+
"train_steps_per_second": 0.067
|
662 |
}
|
663 |
],
|
664 |
"logging_steps": 50,
|