Upload seed_1007
Browse files
sft-n1000/seed_1007/config.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"feat_dim": 601,
|
| 3 |
+
"props": [
|
| 4 |
+
"Egc"
|
| 5 |
+
],
|
| 6 |
+
"polybert": "HAYDERphd/polyBERT",
|
| 7 |
+
"results": {
|
| 8 |
+
"baselines": {
|
| 9 |
+
"single_step": {
|
| 10 |
+
"greedy_oracle": {
|
| 11 |
+
"n": 112,
|
| 12 |
+
"success": 0.7321428571428571,
|
| 13 |
+
"regret": 0.31967715220266063
|
| 14 |
+
},
|
| 15 |
+
"greedy_verifier": {
|
| 16 |
+
"n": 112,
|
| 17 |
+
"success": 0.5714285714285714,
|
| 18 |
+
"regret": 0.4323648955720346
|
| 19 |
+
},
|
| 20 |
+
"random": {
|
| 21 |
+
"n": 112,
|
| 22 |
+
"success": 0.10714285714285714,
|
| 23 |
+
"regret": 1.9536036647172412
|
| 24 |
+
}
|
| 25 |
+
},
|
| 26 |
+
"multi_step": {
|
| 27 |
+
"greedy_oracle": {
|
| 28 |
+
"n": 36,
|
| 29 |
+
"success": 0.16666666666666666,
|
| 30 |
+
"regret": 0.9945511401860553
|
| 31 |
+
},
|
| 32 |
+
"greedy_verifier": {
|
| 33 |
+
"n": 36,
|
| 34 |
+
"success": 0.08333333333333333,
|
| 35 |
+
"regret": 1.2064229705924578
|
| 36 |
+
},
|
| 37 |
+
"random": {
|
| 38 |
+
"n": 36,
|
| 39 |
+
"success": 0.1111111111111111,
|
| 40 |
+
"regret": 2.30599516517197
|
| 41 |
+
}
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"sft": {
|
| 45 |
+
"single_step": {
|
| 46 |
+
"n": 112,
|
| 47 |
+
"success": 0.9464285714285714,
|
| 48 |
+
"regret": 0.034604044261307886
|
| 49 |
+
},
|
| 50 |
+
"multi_step": {
|
| 51 |
+
"n": 36,
|
| 52 |
+
"success": 0.9444444444444444,
|
| 53 |
+
"regret": 0.047631819676513566
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"rl": null,
|
| 57 |
+
"props": [
|
| 58 |
+
"Egc"
|
| 59 |
+
],
|
| 60 |
+
"verifier": "finetuned_polyBERT",
|
| 61 |
+
"val_rmse": {
|
| 62 |
+
"Egc": 0.486987457705025
|
| 63 |
+
},
|
| 64 |
+
"n_train_req": 1000,
|
| 65 |
+
"n_base_req": 369,
|
| 66 |
+
"n_aug_req": 631,
|
| 67 |
+
"n_steps": 5145,
|
| 68 |
+
"sft_only": true,
|
| 69 |
+
"rl_algorithm": "reinforce",
|
| 70 |
+
"rl_config": {
|
| 71 |
+
"iters": 500,
|
| 72 |
+
"kl_beta": 0.0,
|
| 73 |
+
"grpo_batch": 8,
|
| 74 |
+
"grpo_group_size": 8,
|
| 75 |
+
"grpo_update_epochs": 2,
|
| 76 |
+
"grpo_clip_eps": 0.2
|
| 77 |
+
},
|
| 78 |
+
"seed": 1007,
|
| 79 |
+
"wandb_run": "https://wandb.ai/promotion-kim/polyedit/runs/c4qp9tq7"
|
| 80 |
+
}
|
| 81 |
+
}
|
sft-n1000/seed_1007/policy.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3d721a4d19f5fea78521c23c50441d49028920565dbb212ee757ff8aa1e1d00
|
| 3 |
+
size 883401
|
sft-n1000/seed_1007/sft_policy_bundle.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:565e5a6c4485d0deae0b93f0cbf0510abc0d143f8aaf6e5c303aa21cc9eb9cb0
|
| 3 |
+
size 888985
|
sft-n1000/seed_1007/verifier_Egc_bundle.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dfae65844381eed22152ba80c26563b96d031c01158cdea4a1dfeec7a0e48292
|
| 3 |
+
size 100840835
|