Add files using upload-large-folder tool
Browse files- rl_lavida_video_reversal/results/all_rewards.csv +70 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step10.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step11.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step12.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step13.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step15.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step16.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step17.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step18.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step19.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step2.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step22.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step23.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step24.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step28.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step30.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step5.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step7.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step9.txt +2 -0
- rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train.txt +0 -0
rl_lavida_video_reversal/results/all_rewards.csv
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
step,acc_reward,t_grpo_reward,unbiased_reward,response_length
|
| 2 |
+
1,0.4238,0.2578,0.0562,125.62
|
| 3 |
+
step,acc_reward,t_grpo_reward,unbiased_reward,response_length
|
| 4 |
+
1,0.4238,0.2578,0.0562,125.62
|
| 5 |
+
step,acc_reward,t_grpo_reward,unbiased_reward,response_length
|
| 6 |
+
1,0.4824,0.2766,0.0750,126.07
|
| 7 |
+
step,acc_reward,t_grpo_reward,unbiased_reward,response_length
|
| 8 |
+
1,0.4824,0.2766,0.0750,126.07
|
| 9 |
+
step,acc_reward,t_grpo_reward,unbiased_reward,response_length
|
| 10 |
+
1,0.3965,0.2719,0.0937,124.57
|
| 11 |
+
step,acc_reward,t_grpo_reward,unbiased_reward,response_length
|
| 12 |
+
1,0.3965,0.2719,0.0937,124.57
|
| 13 |
+
2,0.4375,0.2719,0.0937,126.60
|
| 14 |
+
2,0.4375,0.2719,0.0937,126.60
|
| 15 |
+
3,0.5137,0.2766,0.0844,124.65
|
| 16 |
+
3,0.5137,0.2766,0.0844,124.65
|
| 17 |
+
4,0.4590,0.2859,0.0750,126.43
|
| 18 |
+
4,0.4590,0.2859,0.0750,126.43
|
| 19 |
+
5,0.5156,0.2719,0.1031,124.63
|
| 20 |
+
5,0.5156,0.2719,0.1031,124.63
|
| 21 |
+
6,0.3926,0.2672,0.0844,126.23
|
| 22 |
+
6,0.3926,0.2672,0.0844,126.23
|
| 23 |
+
7,0.4531,0.2391,0.0750,125.53
|
| 24 |
+
7,0.4531,0.2391,0.0750,125.53
|
| 25 |
+
8,0.4688,0.2906,0.0844,126.02
|
| 26 |
+
8,0.4688,0.2906,0.0844,126.02
|
| 27 |
+
9,0.4570,0.2578,0.1219,126.00
|
| 28 |
+
9,0.4570,0.2578,0.1219,126.00
|
| 29 |
+
10,0.5078,0.2531,0.0937,125.81
|
| 30 |
+
10,0.5078,0.2531,0.0937,125.81
|
| 31 |
+
11,0.4570,0.2438,0.0469,126.20
|
| 32 |
+
11,0.4570,0.2438,0.0469,126.20
|
| 33 |
+
12,0.3457,0.2672,0.0750,126.02
|
| 34 |
+
12,0.3457,0.2672,0.0750,126.02
|
| 35 |
+
13,0.4570,0.2625,0.1125,123.99
|
| 36 |
+
13,0.4570,0.2625,0.1125,123.99
|
| 37 |
+
14,0.4902,0.2578,0.1219,126.41
|
| 38 |
+
14,0.4902,0.2578,0.1219,126.41
|
| 39 |
+
15,0.4648,0.2719,0.0562,126.29
|
| 40 |
+
15,0.4648,0.2719,0.0562,126.29
|
| 41 |
+
16,0.4609,0.2578,0.0937,127.29
|
| 42 |
+
16,0.4609,0.2578,0.0937,127.29
|
| 43 |
+
17,0.5703,0.2859,0.1219,126.11
|
| 44 |
+
17,0.5703,0.2859,0.1219,126.11
|
| 45 |
+
18,0.5273,0.2672,0.0656,125.13
|
| 46 |
+
18,0.5273,0.2672,0.0656,125.13
|
| 47 |
+
19,0.4551,0.2438,0.0937,125.24
|
| 48 |
+
19,0.4551,0.2438,0.0937,125.24
|
| 49 |
+
20,0.4395,0.2813,0.0937,125.99
|
| 50 |
+
20,0.4395,0.2813,0.0937,125.99
|
| 51 |
+
21,0.3711,0.2766,0.0469,126.30
|
| 52 |
+
21,0.3711,0.2766,0.0469,126.30
|
| 53 |
+
22,0.4434,0.2672,0.1031,126.28
|
| 54 |
+
22,0.4434,0.2672,0.1031,126.28
|
| 55 |
+
23,0.4141,0.2719,0.0656,126.45
|
| 56 |
+
23,0.4141,0.2719,0.0656,126.45
|
| 57 |
+
24,0.4551,0.2625,0.1031,124.63
|
| 58 |
+
24,0.4551,0.2625,0.1031,124.63
|
| 59 |
+
25,0.3730,0.2625,0.0750,125.20
|
| 60 |
+
25,0.3730,0.2625,0.0750,125.20
|
| 61 |
+
26,0.4141,0.2719,0.1125,125.16
|
| 62 |
+
26,0.4141,0.2719,0.1125,125.16
|
| 63 |
+
27,0.4609,0.2672,0.1219,125.97
|
| 64 |
+
27,0.4609,0.2672,0.1219,125.97
|
| 65 |
+
28,0.4238,0.2672,0.0750,125.67
|
| 66 |
+
28,0.4238,0.2672,0.0750,125.67
|
| 67 |
+
29,0.5078,0.2906,0.1031,126.24
|
| 68 |
+
29,0.5078,0.2906,0.1031,126.24
|
| 69 |
+
30,0.4473,0.2531,0.0937,125.82
|
| 70 |
+
30,0.4473,0.2531,0.0937,125.82
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step10.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 10 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.5078 avg length: 125.81
|
| 2 |
+
train step: 10 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.5078 avg length: 125.81
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step11.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 11 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4570 avg length: 126.20
|
| 2 |
+
train step: 11 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4570 avg length: 126.20
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step12.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 12 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.3457 avg length: 126.02
|
| 2 |
+
train step: 12 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.3457 avg length: 126.02
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step13.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 13 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4570 avg length: 123.99
|
| 2 |
+
train step: 13 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4570 avg length: 123.99
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step15.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 15 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4648 avg length: 126.29
|
| 2 |
+
train step: 15 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4648 avg length: 126.29
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step16.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 16 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4609 avg length: 127.29
|
| 2 |
+
train step: 16 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4609 avg length: 127.29
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step17.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 17 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.5703 avg length: 126.11
|
| 2 |
+
train step: 17 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.5703 avg length: 126.11
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step18.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 18 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.5273 avg length: 125.13
|
| 2 |
+
train step: 18 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.5273 avg length: 125.13
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step19.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 19 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4551 avg length: 125.24
|
| 2 |
+
train step: 19 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4551 avg length: 125.24
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step2.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 2 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4375 avg length: 126.60
|
| 2 |
+
train step: 2 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4375 avg length: 126.60
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step22.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 22 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4434 avg length: 126.28
|
| 2 |
+
train step: 22 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4434 avg length: 126.28
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step23.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 23 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4141 avg length: 126.45
|
| 2 |
+
train step: 23 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4141 avg length: 126.45
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step24.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 24 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4551 avg length: 124.63
|
| 2 |
+
train step: 24 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4551 avg length: 124.63
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step28.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 28 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4238 avg length: 125.67
|
| 2 |
+
train step: 28 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4238 avg length: 125.67
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step30.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 30 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4473 avg length: 125.82
|
| 2 |
+
train step: 30 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4473 avg length: 125.82
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step5.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 5 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.5156 avg length: 124.63
|
| 2 |
+
train step: 5 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.5156 avg length: 124.63
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step7.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 7 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4531 avg length: 125.53
|
| 2 |
+
train step: 7 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4531 avg length: 125.53
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train-step9.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train step: 9 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4570 avg length: 126.00
|
| 2 |
+
train step: 9 remasking_strategy: low_confidence_dynamic block_size: 32 acc: 0.4570 avg length: 126.00
|
rl_lavida_video_reversal/results/results-rl-...rl_lavida_video_reversal.ckpt.optimized-video_reversal_rl_train.txt
ADDED
|
File without changes
|