End of training
Browse files- README.md +5 -4
- adapter_config.json +39 -0
- adapter_model.safetensors +3 -0
- all_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +118 -118
- training_args.bin +2 -2
README.md
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
---
|
2 |
-
base_model: Gensyn/Qwen2.5-
|
3 |
library_name: transformers
|
4 |
model_name: gensyn-checkpoints-purring_jumping_bear
|
5 |
tags:
|
@@ -8,13 +8,14 @@ tags:
|
|
8 |
- grpo
|
9 |
- gensyn
|
10 |
- I am purring jumping bear
|
|
|
11 |
- trl
|
12 |
licence: license
|
13 |
---
|
14 |
|
15 |
# Model Card for gensyn-checkpoints-purring_jumping_bear
|
16 |
|
17 |
-
This model is a fine-tuned version of [Gensyn/Qwen2.5-
|
18 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
19 |
|
20 |
## Quick start
|
@@ -39,8 +40,8 @@ This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing
|
|
39 |
|
40 |
- TRL: 0.15.2
|
41 |
- Transformers: 4.51.3
|
42 |
-
- Pytorch: 2.
|
43 |
-
- Datasets: 3.
|
44 |
- Tokenizers: 0.21.1
|
45 |
|
46 |
## Citations
|
|
|
1 |
---
|
2 |
+
base_model: Gensyn/Qwen2.5-1.5B-Instruct
|
3 |
library_name: transformers
|
4 |
model_name: gensyn-checkpoints-purring_jumping_bear
|
5 |
tags:
|
|
|
8 |
- grpo
|
9 |
- gensyn
|
10 |
- I am purring jumping bear
|
11 |
+
- unsloth
|
12 |
- trl
|
13 |
licence: license
|
14 |
---
|
15 |
|
16 |
# Model Card for gensyn-checkpoints-purring_jumping_bear
|
17 |
|
18 |
+
This model is a fine-tuned version of [Gensyn/Qwen2.5-1.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-1.5B-Instruct).
|
19 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
20 |
|
21 |
## Quick start
|
|
|
40 |
|
41 |
- TRL: 0.15.2
|
42 |
- Transformers: 4.51.3
|
43 |
+
- Pytorch: 2.6.0
|
44 |
+
- Datasets: 3.6.0
|
45 |
- Tokenizers: 0.21.1
|
46 |
|
47 |
## Citations
|
adapter_config.json
ADDED
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "Gensyn/Qwen2.5-1.5B-Instruct",
|
5 |
+
"bias": "none",
|
6 |
+
"corda_config": null,
|
7 |
+
"eva_config": null,
|
8 |
+
"exclude_modules": null,
|
9 |
+
"fan_in_fan_out": false,
|
10 |
+
"inference_mode": true,
|
11 |
+
"init_lora_weights": true,
|
12 |
+
"layer_replication": null,
|
13 |
+
"layers_pattern": null,
|
14 |
+
"layers_to_transform": null,
|
15 |
+
"loftq_config": {},
|
16 |
+
"lora_alpha": 16,
|
17 |
+
"lora_bias": false,
|
18 |
+
"lora_dropout": 0,
|
19 |
+
"megatron_config": null,
|
20 |
+
"megatron_core": "megatron.core",
|
21 |
+
"modules_to_save": null,
|
22 |
+
"peft_type": "LORA",
|
23 |
+
"r": 16,
|
24 |
+
"rank_pattern": {},
|
25 |
+
"revision": null,
|
26 |
+
"target_modules": [
|
27 |
+
"q_proj",
|
28 |
+
"v_proj",
|
29 |
+
"k_proj",
|
30 |
+
"o_proj",
|
31 |
+
"gate_proj",
|
32 |
+
"up_proj",
|
33 |
+
"down_proj"
|
34 |
+
],
|
35 |
+
"task_type": "CAUSAL_LM",
|
36 |
+
"trainable_token_indices": null,
|
37 |
+
"use_dora": false,
|
38 |
+
"use_rslora": false
|
39 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:076bad1d1ad90a3408af3038dcb53d8424782f8dac2f21420050f842fcb4f033
|
3 |
+
size 73911112
|
all_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"total_flos": 0.0,
|
3 |
-
"train_loss":
|
4 |
-
"train_runtime":
|
5 |
-
"train_samples":
|
6 |
-
"train_samples_per_second": 0.
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
"total_flos": 0.0,
|
3 |
+
"train_loss": 7.104203099572714e-07,
|
4 |
+
"train_runtime": 1721.2094,
|
5 |
+
"train_samples": 100,
|
6 |
+
"train_samples_per_second": 0.186,
|
7 |
+
"train_steps_per_second": 0.012
|
8 |
}
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"total_flos": 0.0,
|
3 |
-
"train_loss":
|
4 |
-
"train_runtime":
|
5 |
-
"train_samples":
|
6 |
-
"train_samples_per_second": 0.
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
"total_flos": 0.0,
|
3 |
+
"train_loss": 7.104203099572714e-07,
|
4 |
+
"train_runtime": 1721.2094,
|
5 |
+
"train_samples": 100,
|
6 |
+
"train_samples_per_second": 0.186,
|
7 |
+
"train_steps_per_second": 0.012
|
8 |
}
|
trainer_state.json
CHANGED
@@ -2,7 +2,7 @@
|
|
2 |
"best_global_step": null,
|
3 |
"best_metric": null,
|
4 |
"best_model_checkpoint": null,
|
5 |
-
"epoch":
|
6 |
"eval_steps": 500,
|
7 |
"global_step": 20,
|
8 |
"is_hyper_param_search": false,
|
@@ -10,209 +10,209 @@
|
|
10 |
"is_world_process_zero": true,
|
11 |
"log_history": [
|
12 |
{
|
13 |
-
"completion_length":
|
14 |
-
"epoch": 0.
|
15 |
-
"grad_norm":
|
16 |
"kl": 0.0,
|
17 |
"learning_rate": 5e-07,
|
18 |
-
"loss": 0.0,
|
19 |
-
"reward":
|
20 |
-
"reward_std":
|
21 |
-
"rewards/concensus_correctness_reward_func": 0.
|
22 |
-
"rewards/consensus_reward_func": 0.
|
23 |
"rewards/cumulative_reward_2": 0.0,
|
24 |
-
"rewards/final_correctness_reward_func": 0.
|
25 |
-
"rewards/question_recreation_reward_func": 0.
|
26 |
"rewards/soft_format_reward_func": 0.0,
|
27 |
-
"rewards/strict_format_reward_func": 0.
|
28 |
-
"rewards/xmlcount_reward_func": 0.
|
29 |
"step": 2
|
30 |
},
|
31 |
{
|
32 |
-
"completion_length":
|
33 |
-
"epoch":
|
34 |
-
"grad_norm":
|
35 |
-
"kl": 0.
|
36 |
"learning_rate": 4.864543104251586e-07,
|
37 |
"loss": 0.0,
|
38 |
-
"reward": 0.
|
39 |
-
"reward_std": 0.
|
40 |
-
"rewards/concensus_correctness_reward_func": 0.
|
41 |
-
"rewards/consensus_reward_func": 0.
|
42 |
"rewards/cumulative_reward_2": 0.0,
|
43 |
"rewards/final_correctness_reward_func": 0.0,
|
44 |
-
"rewards/question_recreation_reward_func": 0.
|
45 |
"rewards/soft_format_reward_func": 0.0,
|
46 |
"rewards/strict_format_reward_func": 0.0,
|
47 |
-
"rewards/xmlcount_reward_func": 0.
|
48 |
"step": 4
|
49 |
},
|
50 |
{
|
51 |
-
"completion_length":
|
52 |
-
"epoch":
|
53 |
-
"grad_norm":
|
54 |
-
"kl": 0.
|
55 |
"learning_rate": 4.472851273490984e-07,
|
56 |
"loss": 0.0,
|
57 |
-
"reward": 0.
|
58 |
-
"reward_std": 0.
|
59 |
-
"rewards/concensus_correctness_reward_func": 0.
|
60 |
-
"rewards/consensus_reward_func": 0.
|
61 |
"rewards/cumulative_reward_2": 0.0,
|
62 |
"rewards/final_correctness_reward_func": 0.0,
|
63 |
-
"rewards/question_recreation_reward_func": 0.
|
64 |
"rewards/soft_format_reward_func": 0.0,
|
65 |
"rewards/strict_format_reward_func": 0.0,
|
66 |
-
"rewards/xmlcount_reward_func": 0.
|
67 |
"step": 6
|
68 |
},
|
69 |
{
|
70 |
-
"completion_length":
|
71 |
-
"epoch":
|
72 |
-
"grad_norm":
|
73 |
-
"kl": 0.
|
74 |
"learning_rate": 3.867370395306068e-07,
|
75 |
"loss": 0.0,
|
76 |
-
"reward": 0.
|
77 |
-
"reward_std": 0.
|
78 |
-
"rewards/concensus_correctness_reward_func": 0.
|
79 |
-
"rewards/consensus_reward_func": 0.
|
80 |
"rewards/cumulative_reward_2": 0.0,
|
81 |
"rewards/final_correctness_reward_func": 0.0,
|
82 |
-
"rewards/question_recreation_reward_func": 0.
|
83 |
"rewards/soft_format_reward_func": 0.0,
|
84 |
-
"rewards/strict_format_reward_func": 0.
|
85 |
-
"rewards/xmlcount_reward_func": 0.
|
86 |
"step": 8
|
87 |
},
|
88 |
{
|
89 |
-
"completion_length":
|
90 |
-
"epoch":
|
91 |
-
"grad_norm":
|
92 |
-
"kl": 0.
|
93 |
"learning_rate": 3.1137137178519977e-07,
|
94 |
"loss": 0.0,
|
95 |
-
"reward":
|
96 |
-
"reward_std":
|
97 |
"rewards/concensus_correctness_reward_func": 0.0,
|
98 |
-
"rewards/consensus_reward_func": 0.
|
99 |
"rewards/cumulative_reward_2": 0.0,
|
100 |
-
"rewards/final_correctness_reward_func": 0.
|
101 |
-
"rewards/question_recreation_reward_func": 0.
|
102 |
"rewards/soft_format_reward_func": 0.0,
|
103 |
-
"rewards/strict_format_reward_func": 0.
|
104 |
-
"rewards/xmlcount_reward_func": 0.
|
105 |
"step": 10
|
106 |
},
|
107 |
{
|
108 |
-
"completion_length":
|
109 |
-
"epoch":
|
110 |
-
"grad_norm":
|
111 |
-
"kl": 0.
|
112 |
"learning_rate": 2.2935516363191693e-07,
|
113 |
"loss": 0.0,
|
114 |
-
"reward": 0.
|
115 |
-
"reward_std":
|
116 |
"rewards/concensus_correctness_reward_func": 0.0,
|
117 |
-
"rewards/consensus_reward_func": 0.
|
118 |
"rewards/cumulative_reward_2": 0.0,
|
119 |
-
"rewards/final_correctness_reward_func": 0.
|
120 |
-
"rewards/question_recreation_reward_func": 0.
|
121 |
"rewards/soft_format_reward_func": 0.0,
|
122 |
-
"rewards/strict_format_reward_func": 0.
|
123 |
-
"rewards/xmlcount_reward_func": 0.
|
124 |
"step": 12
|
125 |
},
|
126 |
{
|
127 |
-
"completion_length":
|
128 |
-
"epoch":
|
129 |
-
"grad_norm":
|
130 |
-
"kl": 0.
|
131 |
"learning_rate": 1.4957614383675767e-07,
|
132 |
"loss": 0.0,
|
133 |
-
"reward": 0.
|
134 |
-
"reward_std":
|
135 |
"rewards/concensus_correctness_reward_func": 0.0,
|
136 |
-
"rewards/consensus_reward_func": 0.
|
137 |
"rewards/cumulative_reward_2": 0.0,
|
138 |
"rewards/final_correctness_reward_func": 0.0,
|
139 |
-
"rewards/question_recreation_reward_func": 0.
|
140 |
"rewards/soft_format_reward_func": 0.0,
|
141 |
-
"rewards/strict_format_reward_func": 0.
|
142 |
-
"rewards/xmlcount_reward_func": 0.
|
143 |
"step": 14
|
144 |
},
|
145 |
{
|
146 |
-
"completion_length":
|
147 |
-
"epoch":
|
148 |
-
"grad_norm":
|
149 |
-
"kl": 0.
|
150 |
"learning_rate": 8.067960709356478e-08,
|
151 |
"loss": 0.0,
|
152 |
-
"reward":
|
153 |
-
"reward_std": 0.
|
154 |
-
"rewards/concensus_correctness_reward_func": 0.
|
155 |
-
"rewards/consensus_reward_func": 0.
|
156 |
"rewards/cumulative_reward_2": 0.0,
|
157 |
-
"rewards/final_correctness_reward_func": 0.
|
158 |
-
"rewards/question_recreation_reward_func": 0.
|
159 |
"rewards/soft_format_reward_func": 0.0,
|
160 |
"rewards/strict_format_reward_func": 0.0,
|
161 |
-
"rewards/xmlcount_reward_func":
|
162 |
"step": 16
|
163 |
},
|
164 |
{
|
165 |
-
"completion_length":
|
166 |
-
"epoch":
|
167 |
-
"grad_norm":
|
168 |
-
"kl": 0.
|
169 |
"learning_rate": 3.013156219837776e-08,
|
170 |
"loss": 0.0,
|
171 |
-
"reward":
|
172 |
-
"reward_std":
|
173 |
-
"rewards/concensus_correctness_reward_func": 0.
|
174 |
-
"rewards/consensus_reward_func": 0.
|
175 |
"rewards/cumulative_reward_2": 0.0,
|
176 |
-
"rewards/final_correctness_reward_func": 0.
|
177 |
-
"rewards/question_recreation_reward_func": 0.
|
178 |
"rewards/soft_format_reward_func": 0.0,
|
179 |
-
"rewards/strict_format_reward_func": 0.
|
180 |
-
"rewards/xmlcount_reward_func":
|
181 |
"step": 18
|
182 |
},
|
183 |
{
|
184 |
-
"completion_length":
|
185 |
-
"epoch":
|
186 |
-
"grad_norm":
|
187 |
-
"kl": 0.
|
188 |
"learning_rate": 3.4096741493194193e-09,
|
189 |
"loss": 0.0,
|
190 |
-
"reward":
|
191 |
-
"reward_std":
|
192 |
-
"rewards/concensus_correctness_reward_func": 0.
|
193 |
-
"rewards/consensus_reward_func": 0.
|
194 |
"rewards/cumulative_reward_2": 0.0,
|
195 |
-
"rewards/final_correctness_reward_func": 0.
|
196 |
-
"rewards/question_recreation_reward_func": 0.
|
197 |
"rewards/soft_format_reward_func": 0.0,
|
198 |
-
"rewards/strict_format_reward_func": 0.
|
199 |
-
"rewards/xmlcount_reward_func": 0.
|
200 |
"step": 20
|
201 |
},
|
202 |
{
|
203 |
-
"epoch":
|
204 |
"step": 20,
|
205 |
"total_flos": 0.0,
|
206 |
-
"train_loss":
|
207 |
-
"train_runtime":
|
208 |
-
"train_samples_per_second": 0.
|
209 |
-
"train_steps_per_second": 0.
|
210 |
}
|
211 |
],
|
212 |
"logging_steps": 2,
|
213 |
"max_steps": 20,
|
214 |
"num_input_tokens_seen": 0,
|
215 |
-
"num_train_epochs":
|
216 |
"save_steps": 25,
|
217 |
"stateful_callbacks": {
|
218 |
"TrainerControl": {
|
@@ -227,7 +227,7 @@
|
|
227 |
}
|
228 |
},
|
229 |
"total_flos": 0.0,
|
230 |
-
"train_batch_size":
|
231 |
"trial_name": null,
|
232 |
"trial_params": null
|
233 |
}
|
|
|
2 |
"best_global_step": null,
|
3 |
"best_metric": null,
|
4 |
"best_model_checkpoint": null,
|
5 |
+
"epoch": 0.8,
|
6 |
"eval_steps": 500,
|
7 |
"global_step": 20,
|
8 |
"is_hyper_param_search": false,
|
|
|
10 |
"is_world_process_zero": true,
|
11 |
"log_history": [
|
12 |
{
|
13 |
+
"completion_length": 473.9375,
|
14 |
+
"epoch": 0.08,
|
15 |
+
"grad_norm": 0.34969469904899597,
|
16 |
"kl": 0.0,
|
17 |
"learning_rate": 5e-07,
|
18 |
+
"loss": -0.0,
|
19 |
+
"reward": 1.483535647392273,
|
20 |
+
"reward_std": 1.2373702749609947,
|
21 |
+
"rewards/concensus_correctness_reward_func": 0.006062500178813934,
|
22 |
+
"rewards/consensus_reward_func": 0.5,
|
23 |
"rewards/cumulative_reward_2": 0.0,
|
24 |
+
"rewards/final_correctness_reward_func": 0.0,
|
25 |
+
"rewards/question_recreation_reward_func": 0.5246606338769197,
|
26 |
"rewards/soft_format_reward_func": 0.0,
|
27 |
+
"rewards/strict_format_reward_func": 0.015625,
|
28 |
+
"rewards/xmlcount_reward_func": 0.4371874928474426,
|
29 |
"step": 2
|
30 |
},
|
31 |
{
|
32 |
+
"completion_length": 466.75,
|
33 |
+
"epoch": 0.16,
|
34 |
+
"grad_norm": 0.31218668818473816,
|
35 |
+
"kl": 0.0007654730725334957,
|
36 |
"learning_rate": 4.864543104251586e-07,
|
37 |
"loss": 0.0,
|
38 |
+
"reward": 0.7086614593863487,
|
39 |
+
"reward_std": 0.8326445110142231,
|
40 |
+
"rewards/concensus_correctness_reward_func": 0.06012500077486038,
|
41 |
+
"rewards/consensus_reward_func": 0.25,
|
42 |
"rewards/cumulative_reward_2": 0.0,
|
43 |
"rewards/final_correctness_reward_func": 0.0,
|
44 |
+
"rewards/question_recreation_reward_func": 0.5004427116364241,
|
45 |
"rewards/soft_format_reward_func": 0.0,
|
46 |
"rewards/strict_format_reward_func": 0.0,
|
47 |
+
"rewards/xmlcount_reward_func": -0.10190625116229057,
|
48 |
"step": 4
|
49 |
},
|
50 |
{
|
51 |
+
"completion_length": 488.75,
|
52 |
+
"epoch": 0.24,
|
53 |
+
"grad_norm": 0.3539212942123413,
|
54 |
+
"kl": 0.0007989688565430697,
|
55 |
"learning_rate": 4.472851273490984e-07,
|
56 |
"loss": 0.0,
|
57 |
+
"reward": 0.9597348868846893,
|
58 |
+
"reward_std": 0.8448215201497078,
|
59 |
+
"rewards/concensus_correctness_reward_func": 0.054749999195337296,
|
60 |
+
"rewards/consensus_reward_func": 0.0625,
|
61 |
"rewards/cumulative_reward_2": 0.0,
|
62 |
"rewards/final_correctness_reward_func": 0.0,
|
63 |
+
"rewards/question_recreation_reward_func": 0.5574536551721394,
|
64 |
"rewards/soft_format_reward_func": 0.0,
|
65 |
"rewards/strict_format_reward_func": 0.0,
|
66 |
+
"rewards/xmlcount_reward_func": 0.2850312504451722,
|
67 |
"step": 6
|
68 |
},
|
69 |
{
|
70 |
+
"completion_length": 429.4375,
|
71 |
+
"epoch": 0.32,
|
72 |
+
"grad_norm": 0.5844979286193848,
|
73 |
+
"kl": 0.000885855064552743,
|
74 |
"learning_rate": 3.867370395306068e-07,
|
75 |
"loss": 0.0,
|
76 |
+
"reward": 0.6777178943157196,
|
77 |
+
"reward_std": 0.7672723643481731,
|
78 |
+
"rewards/concensus_correctness_reward_func": 0.0,
|
79 |
+
"rewards/consensus_reward_func": 0.0625,
|
80 |
"rewards/cumulative_reward_2": 0.0,
|
81 |
"rewards/final_correctness_reward_func": 0.0,
|
82 |
+
"rewards/question_recreation_reward_func": 0.3103741565719247,
|
83 |
"rewards/soft_format_reward_func": 0.0,
|
84 |
+
"rewards/strict_format_reward_func": 0.015625,
|
85 |
+
"rewards/xmlcount_reward_func": 0.28921874798834324,
|
86 |
"step": 8
|
87 |
},
|
88 |
{
|
89 |
+
"completion_length": 436.5625,
|
90 |
+
"epoch": 0.4,
|
91 |
+
"grad_norm": 0.5599504113197327,
|
92 |
+
"kl": 0.0008282829003292136,
|
93 |
"learning_rate": 3.1137137178519977e-07,
|
94 |
"loss": 0.0,
|
95 |
+
"reward": 1.1772124115377665,
|
96 |
+
"reward_std": 1.0518584689125419,
|
97 |
"rewards/concensus_correctness_reward_func": 0.0,
|
98 |
+
"rewards/consensus_reward_func": 0.1875,
|
99 |
"rewards/cumulative_reward_2": 0.0,
|
100 |
+
"rewards/final_correctness_reward_func": 0.125,
|
101 |
+
"rewards/question_recreation_reward_func": 0.5402123816311359,
|
102 |
"rewards/soft_format_reward_func": 0.0,
|
103 |
+
"rewards/strict_format_reward_func": 0.015625,
|
104 |
+
"rewards/xmlcount_reward_func": 0.3088750019669533,
|
105 |
"step": 10
|
106 |
},
|
107 |
{
|
108 |
+
"completion_length": 476.3125,
|
109 |
+
"epoch": 0.48,
|
110 |
+
"grad_norm": 1.196959376335144,
|
111 |
+
"kl": 0.0007865805309847929,
|
112 |
"learning_rate": 2.2935516363191693e-07,
|
113 |
"loss": 0.0,
|
114 |
+
"reward": 0.3430284606292844,
|
115 |
+
"reward_std": 1.131084494292736,
|
116 |
"rewards/concensus_correctness_reward_func": 0.0,
|
117 |
+
"rewards/consensus_reward_func": 0.0625,
|
118 |
"rewards/cumulative_reward_2": 0.0,
|
119 |
+
"rewards/final_correctness_reward_func": 0.0,
|
120 |
+
"rewards/question_recreation_reward_func": 0.3210597266443074,
|
121 |
"rewards/soft_format_reward_func": 0.0,
|
122 |
+
"rewards/strict_format_reward_func": 0.015625,
|
123 |
+
"rewards/xmlcount_reward_func": -0.05615623015910387,
|
124 |
"step": 12
|
125 |
},
|
126 |
{
|
127 |
+
"completion_length": 491.0625,
|
128 |
+
"epoch": 0.56,
|
129 |
+
"grad_norm": 0.3591214418411255,
|
130 |
+
"kl": 0.0007238566759042442,
|
131 |
"learning_rate": 1.4957614383675767e-07,
|
132 |
"loss": 0.0,
|
133 |
+
"reward": 0.9437421336770058,
|
134 |
+
"reward_std": 1.0063903480768204,
|
135 |
"rewards/concensus_correctness_reward_func": 0.0,
|
136 |
+
"rewards/consensus_reward_func": 0.3125,
|
137 |
"rewards/cumulative_reward_2": 0.0,
|
138 |
"rewards/final_correctness_reward_func": 0.0,
|
139 |
+
"rewards/question_recreation_reward_func": 0.6040233373641968,
|
140 |
"rewards/soft_format_reward_func": 0.0,
|
141 |
+
"rewards/strict_format_reward_func": 0.015625,
|
142 |
+
"rewards/xmlcount_reward_func": 0.011593742296099663,
|
143 |
"step": 14
|
144 |
},
|
145 |
{
|
146 |
+
"completion_length": 481.4375,
|
147 |
+
"epoch": 0.64,
|
148 |
+
"grad_norm": 0.5718751549720764,
|
149 |
+
"kl": 0.0007593812551931478,
|
150 |
"learning_rate": 8.067960709356478e-08,
|
151 |
"loss": 0.0,
|
152 |
+
"reward": 1.145691767334938,
|
153 |
+
"reward_std": 0.8273868970572948,
|
154 |
+
"rewards/concensus_correctness_reward_func": 0.012000000104308128,
|
155 |
+
"rewards/consensus_reward_func": 0.0625,
|
156 |
"rewards/cumulative_reward_2": 0.0,
|
157 |
+
"rewards/final_correctness_reward_func": 0.0625,
|
158 |
+
"rewards/question_recreation_reward_func": 0.5370667604729533,
|
159 |
"rewards/soft_format_reward_func": 0.0,
|
160 |
"rewards/strict_format_reward_func": 0.0,
|
161 |
+
"rewards/xmlcount_reward_func": 0.471624992787838,
|
162 |
"step": 16
|
163 |
},
|
164 |
{
|
165 |
+
"completion_length": 464.34375,
|
166 |
+
"epoch": 0.72,
|
167 |
+
"grad_norm": 0.41601094603538513,
|
168 |
+
"kl": 0.0008069156974670477,
|
169 |
"learning_rate": 3.013156219837776e-08,
|
170 |
"loss": 0.0,
|
171 |
+
"reward": 1.989693470299244,
|
172 |
+
"reward_std": 1.9797653332352638,
|
173 |
+
"rewards/concensus_correctness_reward_func": 0.625,
|
174 |
+
"rewards/consensus_reward_func": 0.3125,
|
175 |
"rewards/cumulative_reward_2": 0.0,
|
176 |
+
"rewards/final_correctness_reward_func": 0.25,
|
177 |
+
"rewards/question_recreation_reward_func": 0.5563809499144554,
|
178 |
"rewards/soft_format_reward_func": 0.0,
|
179 |
+
"rewards/strict_format_reward_func": 0.015625,
|
180 |
+
"rewards/xmlcount_reward_func": 0.23018749989569187,
|
181 |
"step": 18
|
182 |
},
|
183 |
{
|
184 |
+
"completion_length": 481.75,
|
185 |
+
"epoch": 0.8,
|
186 |
+
"grad_norm": 0.4525759518146515,
|
187 |
+
"kl": 0.0008362083317479119,
|
188 |
"learning_rate": 3.4096741493194193e-09,
|
189 |
"loss": 0.0,
|
190 |
+
"reward": 1.3046186864376068,
|
191 |
+
"reward_std": 1.2233120650053024,
|
192 |
+
"rewards/concensus_correctness_reward_func": 0.013875000178813934,
|
193 |
+
"rewards/consensus_reward_func": 0.3125,
|
194 |
"rewards/cumulative_reward_2": 0.0,
|
195 |
+
"rewards/final_correctness_reward_func": 0.0625,
|
196 |
+
"rewards/question_recreation_reward_func": 0.6678999401628971,
|
197 |
"rewards/soft_format_reward_func": 0.0,
|
198 |
+
"rewards/strict_format_reward_func": 0.015625,
|
199 |
+
"rewards/xmlcount_reward_func": 0.23221874982118607,
|
200 |
"step": 20
|
201 |
},
|
202 |
{
|
203 |
+
"epoch": 0.8,
|
204 |
"step": 20,
|
205 |
"total_flos": 0.0,
|
206 |
+
"train_loss": 7.104203099572714e-07,
|
207 |
+
"train_runtime": 1721.2094,
|
208 |
+
"train_samples_per_second": 0.186,
|
209 |
+
"train_steps_per_second": 0.012
|
210 |
}
|
211 |
],
|
212 |
"logging_steps": 2,
|
213 |
"max_steps": 20,
|
214 |
"num_input_tokens_seen": 0,
|
215 |
+
"num_train_epochs": 1,
|
216 |
"save_steps": 25,
|
217 |
"stateful_callbacks": {
|
218 |
"TrainerControl": {
|
|
|
227 |
}
|
228 |
},
|
229 |
"total_flos": 0.0,
|
230 |
+
"train_batch_size": 4,
|
231 |
"trial_name": null,
|
232 |
"trial_params": null
|
233 |
}
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6eddf13391de541790c176fc7331830e211a0c4f6babb2091a46b97c829559a8
|
3 |
+
size 5944
|