Upload folder using huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- results/baseline-fulle-lm-wsd-from-1152/zero_shot_eval/checkpoint-1224-backup.json +527 -0
- results/baseline-fulle-lm-wsd-from-1152/zero_shot_eval/checkpoint-1224.json +43 -10
- results/baseline-fulle-lm-wsd-from-1152/zero_shot_eval/summary.jsonl +3 -0
- results/i5-exp-fulle-lm-wsd-from-4320/zero_shot_eval/checkpoint-4464.json +211 -29
- results/i5-exp-fulle-lm-wsd-from-4320/zero_shot_eval/summary.jsonl +2 -0
- results/l2r50-fulle-lm/zero_shot_eval/checkpoint-2304.json +116 -20
- results/l2r50-fulle-lm/zero_shot_eval/summary.jsonl +2 -0
- results/l2r50-i5-exp-fulle-lm-wsd-from-3168/zero_shot_eval/checkpoint-3312.json +211 -29
- results/l2r50-i5-exp-fulle-lm-wsd-from-3168/zero_shot_eval/summary.jsonl +2 -0
- results/random15-fulle-lm-wsd-from-2016/zero_shot_eval/checkpoint-2088.json +211 -29
- results/random15-fulle-lm-wsd-from-2016/zero_shot_eval/summary.jsonl +2 -0
- results/random15-i5-exp-fulle-lm-wsd-from-3168/zero_shot_eval/checkpoint-3672.json +210 -28
- results/random15-i5-exp-fulle-lm-wsd-from-3168/zero_shot_eval/summary.jsonl +2 -0
- results/random15-l2r50-fulle-lm-wsd-from-7488/zero_shot_eval/checkpoint-7776.json +211 -29
- results/random15-l2r50-fulle-lm-wsd-from-7488/zero_shot_eval/summary.jsonl +2 -0
- results/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024/zero_shot_eval/checkpoint-21960.json +211 -29
- results/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024/zero_shot_eval/summary.jsonl +2 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-1152.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-1440.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-1728.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-2016.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-2304.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-2592.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-288.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-2880.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-3168.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-3456.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-3744.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-4032.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-4320.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-4608.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-4896.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-5184.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-5472.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-576.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-5760.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-6048.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-6336.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-6624.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-6912.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-72.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-7200.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/checkpoint-864.json +35 -0
- results/random5-fulle-lm/l2r_n1_eval/summary.jsonl +26 -0
- results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-4968.json +35 -0
- results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-5040.json +35 -0
- results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-5112.json +35 -0
- results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-5184.json +35 -0
- results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-5256.json +35 -0
- results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-5328.json +35 -0
results/baseline-fulle-lm-wsd-from-1152/zero_shot_eval/checkpoint-1224-backup.json
ADDED
|
@@ -0,0 +1,527 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152",
|
| 3 |
+
"run": "baseline-fulle-lm-wsd-from-1152",
|
| 4 |
+
"checkpoint_path": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152/checkpoint-1224",
|
| 5 |
+
"checkpoint": "checkpoint-1224",
|
| 6 |
+
"global_step": 1224,
|
| 7 |
+
"tokens_seen": 1283457024,
|
| 8 |
+
"num_fewshot": 0,
|
| 9 |
+
"runs_per_task": 5,
|
| 10 |
+
"base_seed": 42,
|
| 11 |
+
"tasks_evaluated": [
|
| 12 |
+
"arc_challenge",
|
| 13 |
+
"arc_easy",
|
| 14 |
+
"boolq",
|
| 15 |
+
"copa",
|
| 16 |
+
"hellaswag",
|
| 17 |
+
"lambada_openai",
|
| 18 |
+
"openbookqa",
|
| 19 |
+
"piqa",
|
| 20 |
+
"rte",
|
| 21 |
+
"sciq",
|
| 22 |
+
"winogrande"
|
| 23 |
+
],
|
| 24 |
+
"tasks": {
|
| 25 |
+
"hellaswag": {
|
| 26 |
+
"acc_norm": 0.2513443537143995,
|
| 27 |
+
"acc_norm_ci": NaN,
|
| 28 |
+
"acc_norm_runs": [
|
| 29 |
+
0.2513443537143995,
|
| 30 |
+
0.2513443537143995,
|
| 31 |
+
0.2513443537143995,
|
| 32 |
+
0.2513443537143995,
|
| 33 |
+
0.2513443537143995
|
| 34 |
+
],
|
| 35 |
+
"acc": 0.25761800438159727,
|
| 36 |
+
"acc_ci": NaN,
|
| 37 |
+
"acc_runs": [
|
| 38 |
+
0.25761800438159727,
|
| 39 |
+
0.25761800438159727,
|
| 40 |
+
0.25761800438159727,
|
| 41 |
+
0.25761800438159727,
|
| 42 |
+
0.25761800438159727
|
| 43 |
+
],
|
| 44 |
+
"perplexity": null,
|
| 45 |
+
"perplexity_ci": 0.0,
|
| 46 |
+
"perplexity_runs": [],
|
| 47 |
+
"stderr": 0.0043289955103121775,
|
| 48 |
+
"centered_acc": 0.001792471619199339,
|
| 49 |
+
"centered_acc_ci": NaN,
|
| 50 |
+
"centered_acc_runs": [
|
| 51 |
+
0.001792471619199339,
|
| 52 |
+
0.001792471619199339,
|
| 53 |
+
0.001792471619199339,
|
| 54 |
+
0.001792471619199339,
|
| 55 |
+
0.001792471619199339
|
| 56 |
+
],
|
| 57 |
+
"ci": NaN
|
| 58 |
+
},
|
| 59 |
+
"piqa": {
|
| 60 |
+
"acc_norm": 0.5603917301414582,
|
| 61 |
+
"acc_norm_ci": NaN,
|
| 62 |
+
"acc_norm_runs": [
|
| 63 |
+
0.5603917301414582,
|
| 64 |
+
0.5603917301414582,
|
| 65 |
+
0.5603917301414582,
|
| 66 |
+
0.5603917301414582,
|
| 67 |
+
0.5603917301414582
|
| 68 |
+
],
|
| 69 |
+
"acc": 0.5680087051142546,
|
| 70 |
+
"acc_ci": NaN,
|
| 71 |
+
"acc_runs": [
|
| 72 |
+
0.5680087051142546,
|
| 73 |
+
0.5680087051142546,
|
| 74 |
+
0.5680087051142546,
|
| 75 |
+
0.5680087051142546,
|
| 76 |
+
0.5680087051142546
|
| 77 |
+
],
|
| 78 |
+
"perplexity": null,
|
| 79 |
+
"perplexity_ci": 0.0,
|
| 80 |
+
"perplexity_runs": [],
|
| 81 |
+
"stderr": 0.011580417248656388,
|
| 82 |
+
"centered_acc": 0.12078346028291631,
|
| 83 |
+
"centered_acc_ci": NaN,
|
| 84 |
+
"centered_acc_runs": [
|
| 85 |
+
0.12078346028291631,
|
| 86 |
+
0.12078346028291631,
|
| 87 |
+
0.12078346028291631,
|
| 88 |
+
0.12078346028291631,
|
| 89 |
+
0.12078346028291631
|
| 90 |
+
],
|
| 91 |
+
"ci": NaN
|
| 92 |
+
},
|
| 93 |
+
"arc_easy": {
|
| 94 |
+
"acc_norm": 0.2760942760942761,
|
| 95 |
+
"acc_norm_ci": NaN,
|
| 96 |
+
"acc_norm_runs": [
|
| 97 |
+
0.2760942760942761,
|
| 98 |
+
0.2760942760942761,
|
| 99 |
+
0.2760942760942761,
|
| 100 |
+
0.2760942760942761,
|
| 101 |
+
0.2760942760942761
|
| 102 |
+
],
|
| 103 |
+
"acc": 0.2609427609427609,
|
| 104 |
+
"acc_ci": NaN,
|
| 105 |
+
"acc_runs": [
|
| 106 |
+
0.2609427609427609,
|
| 107 |
+
0.2609427609427609,
|
| 108 |
+
0.2609427609427609,
|
| 109 |
+
0.2609427609427609,
|
| 110 |
+
0.2609427609427609
|
| 111 |
+
],
|
| 112 |
+
"perplexity": null,
|
| 113 |
+
"perplexity_ci": 0.0,
|
| 114 |
+
"perplexity_runs": [],
|
| 115 |
+
"stderr": 0.00917355987383544,
|
| 116 |
+
"centered_acc": 0.03479236812570149,
|
| 117 |
+
"centered_acc_ci": NaN,
|
| 118 |
+
"centered_acc_runs": [
|
| 119 |
+
0.03479236812570149,
|
| 120 |
+
0.03479236812570149,
|
| 121 |
+
0.03479236812570149,
|
| 122 |
+
0.03479236812570149,
|
| 123 |
+
0.03479236812570149
|
| 124 |
+
],
|
| 125 |
+
"ci": NaN
|
| 126 |
+
},
|
| 127 |
+
"arc_challenge": {
|
| 128 |
+
"acc_norm": 0.24914675767918087,
|
| 129 |
+
"acc_norm_ci": NaN,
|
| 130 |
+
"acc_norm_runs": [
|
| 131 |
+
0.24914675767918087,
|
| 132 |
+
0.24914675767918087,
|
| 133 |
+
0.24914675767918087,
|
| 134 |
+
0.24914675767918087,
|
| 135 |
+
0.24914675767918087
|
| 136 |
+
],
|
| 137 |
+
"acc": 0.20648464163822525,
|
| 138 |
+
"acc_ci": NaN,
|
| 139 |
+
"acc_runs": [
|
| 140 |
+
0.20648464163822525,
|
| 141 |
+
0.20648464163822525,
|
| 142 |
+
0.20648464163822525,
|
| 143 |
+
0.20648464163822525,
|
| 144 |
+
0.20648464163822525
|
| 145 |
+
],
|
| 146 |
+
"perplexity": null,
|
| 147 |
+
"perplexity_ci": 0.0,
|
| 148 |
+
"perplexity_runs": [],
|
| 149 |
+
"stderr": 0.012639407111926505,
|
| 150 |
+
"centered_acc": -0.0011376564277588337,
|
| 151 |
+
"centered_acc_ci": NaN,
|
| 152 |
+
"centered_acc_runs": [
|
| 153 |
+
-0.0011376564277588337,
|
| 154 |
+
-0.0011376564277588337,
|
| 155 |
+
-0.0011376564277588337,
|
| 156 |
+
-0.0011376564277588337,
|
| 157 |
+
-0.0011376564277588337
|
| 158 |
+
],
|
| 159 |
+
"ci": NaN
|
| 160 |
+
},
|
| 161 |
+
"openbookqa": {
|
| 162 |
+
"acc_norm": 0.284,
|
| 163 |
+
"acc_norm_ci": NaN,
|
| 164 |
+
"acc_norm_runs": [
|
| 165 |
+
0.284,
|
| 166 |
+
0.284,
|
| 167 |
+
0.284,
|
| 168 |
+
0.284,
|
| 169 |
+
0.284
|
| 170 |
+
],
|
| 171 |
+
"acc": 0.154,
|
| 172 |
+
"acc_ci": NaN,
|
| 173 |
+
"acc_runs": [
|
| 174 |
+
0.154,
|
| 175 |
+
0.154,
|
| 176 |
+
0.154,
|
| 177 |
+
0.154,
|
| 178 |
+
0.154
|
| 179 |
+
],
|
| 180 |
+
"perplexity": null,
|
| 181 |
+
"perplexity_ci": 0.0,
|
| 182 |
+
"perplexity_runs": [],
|
| 183 |
+
"stderr": 0.020186703693570774,
|
| 184 |
+
"centered_acc": 0.0453333333333333,
|
| 185 |
+
"centered_acc_ci": NaN,
|
| 186 |
+
"centered_acc_runs": [
|
| 187 |
+
0.0453333333333333,
|
| 188 |
+
0.0453333333333333,
|
| 189 |
+
0.0453333333333333,
|
| 190 |
+
0.0453333333333333,
|
| 191 |
+
0.0453333333333333
|
| 192 |
+
],
|
| 193 |
+
"ci": NaN
|
| 194 |
+
},
|
| 195 |
+
"winogrande": {
|
| 196 |
+
"acc_norm": null,
|
| 197 |
+
"acc_norm_ci": 0.0,
|
| 198 |
+
"acc_norm_runs": [],
|
| 199 |
+
"acc": 0.48697711128650356,
|
| 200 |
+
"acc_ci": NaN,
|
| 201 |
+
"acc_runs": [
|
| 202 |
+
0.48697711128650356,
|
| 203 |
+
0.48697711128650356,
|
| 204 |
+
0.48697711128650356,
|
| 205 |
+
0.48697711128650356,
|
| 206 |
+
0.48697711128650356
|
| 207 |
+
],
|
| 208 |
+
"perplexity": null,
|
| 209 |
+
"perplexity_ci": 0.0,
|
| 210 |
+
"perplexity_runs": [],
|
| 211 |
+
"stderr": 0.014047718393997596,
|
| 212 |
+
"centered_acc": -0.026045777426992878,
|
| 213 |
+
"centered_acc_ci": NaN,
|
| 214 |
+
"centered_acc_runs": [
|
| 215 |
+
-0.026045777426992878,
|
| 216 |
+
-0.026045777426992878,
|
| 217 |
+
-0.026045777426992878,
|
| 218 |
+
-0.026045777426992878,
|
| 219 |
+
-0.026045777426992878
|
| 220 |
+
],
|
| 221 |
+
"ci": NaN
|
| 222 |
+
},
|
| 223 |
+
"boolq": {
|
| 224 |
+
"acc_norm": null,
|
| 225 |
+
"acc_norm_ci": 0.0,
|
| 226 |
+
"acc_norm_runs": [],
|
| 227 |
+
"acc": 0.3782874617737003,
|
| 228 |
+
"acc_ci": NaN,
|
| 229 |
+
"acc_runs": [
|
| 230 |
+
0.3782874617737003,
|
| 231 |
+
0.3782874617737003,
|
| 232 |
+
0.3782874617737003,
|
| 233 |
+
0.3782874617737003,
|
| 234 |
+
0.3782874617737003
|
| 235 |
+
],
|
| 236 |
+
"perplexity": null,
|
| 237 |
+
"perplexity_ci": 0.0,
|
| 238 |
+
"perplexity_runs": [],
|
| 239 |
+
"stderr": 0.008482001133931097,
|
| 240 |
+
"centered_acc": -0.24342507645259936,
|
| 241 |
+
"centered_acc_ci": NaN,
|
| 242 |
+
"centered_acc_runs": [
|
| 243 |
+
-0.24342507645259936,
|
| 244 |
+
-0.24342507645259936,
|
| 245 |
+
-0.24342507645259936,
|
| 246 |
+
-0.24342507645259936,
|
| 247 |
+
-0.24342507645259936
|
| 248 |
+
],
|
| 249 |
+
"ci": NaN
|
| 250 |
+
},
|
| 251 |
+
"copa": {
|
| 252 |
+
"acc_norm": null,
|
| 253 |
+
"acc_norm_ci": 0.0,
|
| 254 |
+
"acc_norm_runs": [],
|
| 255 |
+
"acc": 0.5,
|
| 256 |
+
"acc_ci": NaN,
|
| 257 |
+
"acc_runs": [
|
| 258 |
+
0.5,
|
| 259 |
+
0.5,
|
| 260 |
+
0.5,
|
| 261 |
+
0.5,
|
| 262 |
+
0.5
|
| 263 |
+
],
|
| 264 |
+
"perplexity": null,
|
| 265 |
+
"perplexity_ci": 0.0,
|
| 266 |
+
"perplexity_runs": [],
|
| 267 |
+
"stderr": 0.050251890762960605,
|
| 268 |
+
"centered_acc": 0.0,
|
| 269 |
+
"centered_acc_ci": NaN,
|
| 270 |
+
"centered_acc_runs": [
|
| 271 |
+
0.0,
|
| 272 |
+
0.0,
|
| 273 |
+
0.0,
|
| 274 |
+
0.0,
|
| 275 |
+
0.0
|
| 276 |
+
],
|
| 277 |
+
"ci": NaN
|
| 278 |
+
},
|
| 279 |
+
"lambada_openai": {
|
| 280 |
+
"acc_norm": null,
|
| 281 |
+
"acc_norm_ci": 0.0,
|
| 282 |
+
"acc_norm_runs": [],
|
| 283 |
+
"acc": 0.0,
|
| 284 |
+
"acc_ci": NaN,
|
| 285 |
+
"acc_runs": [
|
| 286 |
+
0.0,
|
| 287 |
+
0.0,
|
| 288 |
+
0.0,
|
| 289 |
+
0.0,
|
| 290 |
+
0.0
|
| 291 |
+
],
|
| 292 |
+
"perplexity": 391991.009755762,
|
| 293 |
+
"perplexity_ci": NaN,
|
| 294 |
+
"perplexity_runs": [
|
| 295 |
+
391991.009755762,
|
| 296 |
+
391991.009755762,
|
| 297 |
+
391991.009755762,
|
| 298 |
+
391991.009755762,
|
| 299 |
+
391991.009755762
|
| 300 |
+
],
|
| 301 |
+
"stderr": 0.0,
|
| 302 |
+
"centered_acc": null,
|
| 303 |
+
"centered_acc_ci": 0.0,
|
| 304 |
+
"centered_acc_runs": [],
|
| 305 |
+
"ci": NaN
|
| 306 |
+
},
|
| 307 |
+
"sciq": {
|
| 308 |
+
"acc_norm": 0.191,
|
| 309 |
+
"acc_norm_ci": NaN,
|
| 310 |
+
"acc_norm_runs": [
|
| 311 |
+
0.191,
|
| 312 |
+
0.191,
|
| 313 |
+
0.191,
|
| 314 |
+
0.191,
|
| 315 |
+
0.191
|
| 316 |
+
],
|
| 317 |
+
"acc": 0.054000000000000006,
|
| 318 |
+
"acc_ci": 6.938893903907228e-18,
|
| 319 |
+
"acc_runs": [
|
| 320 |
+
0.054,
|
| 321 |
+
0.054,
|
| 322 |
+
0.054,
|
| 323 |
+
0.054,
|
| 324 |
+
0.054
|
| 325 |
+
],
|
| 326 |
+
"perplexity": null,
|
| 327 |
+
"perplexity_ci": 0.0,
|
| 328 |
+
"perplexity_runs": [],
|
| 329 |
+
"stderr": 0.012436787112179482,
|
| 330 |
+
"centered_acc": -0.07866666666666666,
|
| 331 |
+
"centered_acc_ci": NaN,
|
| 332 |
+
"centered_acc_runs": [
|
| 333 |
+
-0.07866666666666666,
|
| 334 |
+
-0.07866666666666666,
|
| 335 |
+
-0.07866666666666666,
|
| 336 |
+
-0.07866666666666666,
|
| 337 |
+
-0.07866666666666666
|
| 338 |
+
],
|
| 339 |
+
"ci": NaN
|
| 340 |
+
},
|
| 341 |
+
"rte": {
|
| 342 |
+
"acc_norm": null,
|
| 343 |
+
"acc_norm_ci": 0.0,
|
| 344 |
+
"acc_norm_runs": [],
|
| 345 |
+
"acc": 0.5270758122743683,
|
| 346 |
+
"acc_ci": NaN,
|
| 347 |
+
"acc_runs": [
|
| 348 |
+
0.5270758122743683,
|
| 349 |
+
0.5270758122743683,
|
| 350 |
+
0.5270758122743683,
|
| 351 |
+
0.5270758122743683,
|
| 352 |
+
0.5270758122743683
|
| 353 |
+
],
|
| 354 |
+
"perplexity": null,
|
| 355 |
+
"perplexity_ci": 0.0,
|
| 356 |
+
"perplexity_runs": [],
|
| 357 |
+
"stderr": 0.030052303463143738,
|
| 358 |
+
"centered_acc": 0.054151624548736566,
|
| 359 |
+
"centered_acc_ci": NaN,
|
| 360 |
+
"centered_acc_runs": [
|
| 361 |
+
0.054151624548736566,
|
| 362 |
+
0.054151624548736566,
|
| 363 |
+
0.054151624548736566,
|
| 364 |
+
0.054151624548736566,
|
| 365 |
+
0.054151624548736566
|
| 366 |
+
],
|
| 367 |
+
"ci": NaN
|
| 368 |
+
},
|
| 369 |
+
"summary": {
|
| 370 |
+
"mean_acc_norm": 0.3019961862715526,
|
| 371 |
+
"mean_acc_norm_ci": 0.04532134416790795,
|
| 372 |
+
"mean_acc": 0.30849040885558277,
|
| 373 |
+
"mean_acc_ci": 0.05121294953603617,
|
| 374 |
+
"mean_centered_acc": -0.009242191906413072,
|
| 375 |
+
"mean_centered_acc_ci": 0.02663212685397144,
|
| 376 |
+
"mean_acc_norm_runs": [
|
| 377 |
+
0.2513443537143995,
|
| 378 |
+
0.2513443537143995,
|
| 379 |
+
0.2513443537143995,
|
| 380 |
+
0.2513443537143995,
|
| 381 |
+
0.2513443537143995,
|
| 382 |
+
0.5603917301414582,
|
| 383 |
+
0.5603917301414582,
|
| 384 |
+
0.5603917301414582,
|
| 385 |
+
0.5603917301414582,
|
| 386 |
+
0.5603917301414582,
|
| 387 |
+
0.2760942760942761,
|
| 388 |
+
0.2760942760942761,
|
| 389 |
+
0.2760942760942761,
|
| 390 |
+
0.2760942760942761,
|
| 391 |
+
0.2760942760942761,
|
| 392 |
+
0.24914675767918087,
|
| 393 |
+
0.24914675767918087,
|
| 394 |
+
0.24914675767918087,
|
| 395 |
+
0.24914675767918087,
|
| 396 |
+
0.24914675767918087,
|
| 397 |
+
0.284,
|
| 398 |
+
0.284,
|
| 399 |
+
0.284,
|
| 400 |
+
0.284,
|
| 401 |
+
0.284,
|
| 402 |
+
0.191,
|
| 403 |
+
0.191,
|
| 404 |
+
0.191,
|
| 405 |
+
0.191,
|
| 406 |
+
0.191
|
| 407 |
+
],
|
| 408 |
+
"mean_acc_runs": [
|
| 409 |
+
0.25761800438159727,
|
| 410 |
+
0.25761800438159727,
|
| 411 |
+
0.25761800438159727,
|
| 412 |
+
0.25761800438159727,
|
| 413 |
+
0.25761800438159727,
|
| 414 |
+
0.5680087051142546,
|
| 415 |
+
0.5680087051142546,
|
| 416 |
+
0.5680087051142546,
|
| 417 |
+
0.5680087051142546,
|
| 418 |
+
0.5680087051142546,
|
| 419 |
+
0.2609427609427609,
|
| 420 |
+
0.2609427609427609,
|
| 421 |
+
0.2609427609427609,
|
| 422 |
+
0.2609427609427609,
|
| 423 |
+
0.2609427609427609,
|
| 424 |
+
0.20648464163822525,
|
| 425 |
+
0.20648464163822525,
|
| 426 |
+
0.20648464163822525,
|
| 427 |
+
0.20648464163822525,
|
| 428 |
+
0.20648464163822525,
|
| 429 |
+
0.154,
|
| 430 |
+
0.154,
|
| 431 |
+
0.154,
|
| 432 |
+
0.154,
|
| 433 |
+
0.154,
|
| 434 |
+
0.48697711128650356,
|
| 435 |
+
0.48697711128650356,
|
| 436 |
+
0.48697711128650356,
|
| 437 |
+
0.48697711128650356,
|
| 438 |
+
0.48697711128650356,
|
| 439 |
+
0.3782874617737003,
|
| 440 |
+
0.3782874617737003,
|
| 441 |
+
0.3782874617737003,
|
| 442 |
+
0.3782874617737003,
|
| 443 |
+
0.3782874617737003,
|
| 444 |
+
0.5,
|
| 445 |
+
0.5,
|
| 446 |
+
0.5,
|
| 447 |
+
0.5,
|
| 448 |
+
0.5,
|
| 449 |
+
0.0,
|
| 450 |
+
0.0,
|
| 451 |
+
0.0,
|
| 452 |
+
0.0,
|
| 453 |
+
0.0,
|
| 454 |
+
0.054,
|
| 455 |
+
0.054,
|
| 456 |
+
0.054,
|
| 457 |
+
0.054,
|
| 458 |
+
0.054,
|
| 459 |
+
0.5270758122743683,
|
| 460 |
+
0.5270758122743683,
|
| 461 |
+
0.5270758122743683,
|
| 462 |
+
0.5270758122743683,
|
| 463 |
+
0.5270758122743683
|
| 464 |
+
],
|
| 465 |
+
"mean_centered_acc_runs": [
|
| 466 |
+
0.001792471619199339,
|
| 467 |
+
0.001792471619199339,
|
| 468 |
+
0.001792471619199339,
|
| 469 |
+
0.001792471619199339,
|
| 470 |
+
0.001792471619199339,
|
| 471 |
+
0.12078346028291631,
|
| 472 |
+
0.12078346028291631,
|
| 473 |
+
0.12078346028291631,
|
| 474 |
+
0.12078346028291631,
|
| 475 |
+
0.12078346028291631,
|
| 476 |
+
0.03479236812570149,
|
| 477 |
+
0.03479236812570149,
|
| 478 |
+
0.03479236812570149,
|
| 479 |
+
0.03479236812570149,
|
| 480 |
+
0.03479236812570149,
|
| 481 |
+
-0.0011376564277588337,
|
| 482 |
+
-0.0011376564277588337,
|
| 483 |
+
-0.0011376564277588337,
|
| 484 |
+
-0.0011376564277588337,
|
| 485 |
+
-0.0011376564277588337,
|
| 486 |
+
0.0453333333333333,
|
| 487 |
+
0.0453333333333333,
|
| 488 |
+
0.0453333333333333,
|
| 489 |
+
0.0453333333333333,
|
| 490 |
+
0.0453333333333333,
|
| 491 |
+
-0.026045777426992878,
|
| 492 |
+
-0.026045777426992878,
|
| 493 |
+
-0.026045777426992878,
|
| 494 |
+
-0.026045777426992878,
|
| 495 |
+
-0.026045777426992878,
|
| 496 |
+
-0.24342507645259936,
|
| 497 |
+
-0.24342507645259936,
|
| 498 |
+
-0.24342507645259936,
|
| 499 |
+
-0.24342507645259936,
|
| 500 |
+
-0.24342507645259936,
|
| 501 |
+
0.0,
|
| 502 |
+
0.0,
|
| 503 |
+
0.0,
|
| 504 |
+
0.0,
|
| 505 |
+
0.0,
|
| 506 |
+
-0.07866666666666666,
|
| 507 |
+
-0.07866666666666666,
|
| 508 |
+
-0.07866666666666666,
|
| 509 |
+
-0.07866666666666666,
|
| 510 |
+
-0.07866666666666666,
|
| 511 |
+
0.054151624548736566,
|
| 512 |
+
0.054151624548736566,
|
| 513 |
+
0.054151624548736566,
|
| 514 |
+
0.054151624548736566,
|
| 515 |
+
0.054151624548736566
|
| 516 |
+
],
|
| 517 |
+
"ci": 0.04532134416790795
|
| 518 |
+
},
|
| 519 |
+
"batch_size_per_device": 8,
|
| 520 |
+
"device": "cuda:0",
|
| 521 |
+
"dtype": "float16",
|
| 522 |
+
"add_l2r_token": false,
|
| 523 |
+
"add_next_1_pred_token": false,
|
| 524 |
+
"tokenizer_has_l2r_pred": true,
|
| 525 |
+
"tokenizer_has_next_1_pred": false,
|
| 526 |
+
"max_next_i_train": 1
|
| 527 |
+
}
|
results/baseline-fulle-lm-wsd-from-1152/zero_shot_eval/checkpoint-1224.json
CHANGED
|
@@ -12,6 +12,7 @@
|
|
| 12 |
"arc_challenge",
|
| 13 |
"arc_easy",
|
| 14 |
"boolq",
|
|
|
|
| 15 |
"copa",
|
| 16 |
"hellaswag",
|
| 17 |
"lambada_openai",
|
|
@@ -289,14 +290,14 @@
|
|
| 289 |
0.0,
|
| 290 |
0.0
|
| 291 |
],
|
| 292 |
-
"perplexity": 391991.
|
| 293 |
"perplexity_ci": NaN,
|
| 294 |
"perplexity_runs": [
|
| 295 |
-
391991.
|
| 296 |
-
391991.
|
| 297 |
-
391991.
|
| 298 |
-
391991.
|
| 299 |
-
391991.
|
| 300 |
],
|
| 301 |
"stderr": 0.0,
|
| 302 |
"centered_acc": null,
|
|
@@ -365,13 +366,35 @@
|
|
| 365 |
0.054151624548736566
|
| 366 |
],
|
| 367 |
"ci": NaN
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 368 |
}
|
| 369 |
},
|
| 370 |
"summary": {
|
| 371 |
"mean_acc_norm": 0.3019961862715526,
|
| 372 |
"mean_acc_norm_ci": 0.04532134416790795,
|
| 373 |
-
"mean_acc": 0.
|
| 374 |
-
"mean_acc_ci": 0.
|
| 375 |
"mean_centered_acc": -0.009242191906413072,
|
| 376 |
"mean_centered_acc_ci": 0.02663212685397144,
|
| 377 |
"mean_acc_norm_runs": [
|
|
@@ -461,7 +484,17 @@
|
|
| 461 |
0.5270758122743683,
|
| 462 |
0.5270758122743683,
|
| 463 |
0.5270758122743683,
|
| 464 |
-
0.5270758122743683
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 465 |
],
|
| 466 |
"mean_centered_acc_runs": [
|
| 467 |
0.001792471619199339,
|
|
@@ -525,4 +558,4 @@
|
|
| 525 |
"tokenizer_has_l2r_pred": true,
|
| 526 |
"tokenizer_has_next_1_pred": false,
|
| 527 |
"max_next_i_train": 1
|
| 528 |
-
}
|
|
|
|
| 12 |
"arc_challenge",
|
| 13 |
"arc_easy",
|
| 14 |
"boolq",
|
| 15 |
+
"commonsense_qa",
|
| 16 |
"copa",
|
| 17 |
"hellaswag",
|
| 18 |
"lambada_openai",
|
|
|
|
| 290 |
0.0,
|
| 291 |
0.0
|
| 292 |
],
|
| 293 |
+
"perplexity": 391991.60405657435,
|
| 294 |
"perplexity_ci": NaN,
|
| 295 |
"perplexity_runs": [
|
| 296 |
+
391991.60405657435,
|
| 297 |
+
391991.60405657435,
|
| 298 |
+
391991.60405657435,
|
| 299 |
+
391991.60405657435,
|
| 300 |
+
391991.60405657435
|
| 301 |
],
|
| 302 |
"stderr": 0.0,
|
| 303 |
"centered_acc": null,
|
|
|
|
| 366 |
0.054151624548736566
|
| 367 |
],
|
| 368 |
"ci": NaN
|
| 369 |
+
},
|
| 370 |
+
"commonsense_qa": {
|
| 371 |
+
"acc_norm": null,
|
| 372 |
+
"acc_norm_ci": 0.0,
|
| 373 |
+
"acc_norm_runs": [],
|
| 374 |
+
"acc": 0.19574119574119575,
|
| 375 |
+
"acc_ci": NaN,
|
| 376 |
+
"acc_runs": [
|
| 377 |
+
0.19574119574119575,
|
| 378 |
+
0.19574119574119575,
|
| 379 |
+
0.19574119574119575,
|
| 380 |
+
0.19574119574119575,
|
| 381 |
+
0.19574119574119575
|
| 382 |
+
],
|
| 383 |
+
"perplexity": null,
|
| 384 |
+
"perplexity_ci": 0.0,
|
| 385 |
+
"perplexity_runs": [],
|
| 386 |
+
"stderr": 0.011359497363584397,
|
| 387 |
+
"centered_acc": null,
|
| 388 |
+
"centered_acc_ci": 0.0,
|
| 389 |
+
"centered_acc_runs": [],
|
| 390 |
+
"ci": NaN
|
| 391 |
}
|
| 392 |
},
|
| 393 |
"summary": {
|
| 394 |
"mean_acc_norm": 0.3019961862715526,
|
| 395 |
"mean_acc_norm_ci": 0.04532134416790795,
|
| 396 |
+
"mean_acc": 0.3261688765111649,
|
| 397 |
+
"mean_acc_ci": 0.04963449075055859,
|
| 398 |
"mean_centered_acc": -0.009242191906413072,
|
| 399 |
"mean_centered_acc_ci": 0.02663212685397144,
|
| 400 |
"mean_acc_norm_runs": [
|
|
|
|
| 484 |
0.5270758122743683,
|
| 485 |
0.5270758122743683,
|
| 486 |
0.5270758122743683,
|
| 487 |
+
0.5270758122743683,
|
| 488 |
+
0.19574119574119575,
|
| 489 |
+
0.19574119574119575,
|
| 490 |
+
0.19574119574119575,
|
| 491 |
+
0.19574119574119575,
|
| 492 |
+
0.19574119574119575,
|
| 493 |
+
0.6510597014925373,
|
| 494 |
+
0.6510597014925373,
|
| 495 |
+
0.6510597014925373,
|
| 496 |
+
0.6510597014925373,
|
| 497 |
+
0.6510597014925373
|
| 498 |
],
|
| 499 |
"mean_centered_acc_runs": [
|
| 500 |
0.001792471619199339,
|
|
|
|
| 558 |
"tokenizer_has_l2r_pred": true,
|
| 559 |
"tokenizer_has_next_1_pred": false,
|
| 560 |
"max_next_i_train": 1
|
| 561 |
+
}
|
results/baseline-fulle-lm-wsd-from-1152/zero_shot_eval/summary.jsonl
CHANGED
|
@@ -1,2 +1,5 @@
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152", "run": "baseline-fulle-lm-wsd-from-1152", "checkpoint_path": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152/checkpoint-1224", "checkpoint": "checkpoint-1224", "global_step": 1224, "tokens_seen": 1283457024, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2513443537143995, "acc_norm_ci": NaN, "acc_norm_runs": [0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995], "acc": 0.25761800438159727, "acc_ci": NaN, "acc_runs": [0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0043289955103121775, "centered_acc": 0.001792471619199339, "centered_acc_ci": NaN, "centered_acc_runs": [0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339], "ci": NaN}, "piqa": {"acc_norm": 0.5603917301414582, "acc_norm_ci": NaN, "acc_norm_runs": [0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582], "acc": 0.5680087051142546, "acc_ci": NaN, "acc_runs": [0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011580417248656388, "centered_acc": 0.12078346028291631, "centered_acc_ci": NaN, "centered_acc_runs": [0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631], "ci": NaN}, "arc_easy": {"acc_norm": 0.2760942760942761, "acc_norm_ci": NaN, "acc_norm_runs": [0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761], "acc": 0.2609427609427609, "acc_ci": NaN, "acc_runs": [0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00917355987383544, "centered_acc": 0.03479236812570149, "centered_acc_ci": NaN, "centered_acc_runs": [0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149], "ci": NaN}, "arc_challenge": {"acc_norm": 0.24914675767918087, "acc_norm_ci": NaN, "acc_norm_runs": [0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087], "acc": 0.20648464163822525, "acc_ci": NaN, "acc_runs": [0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012639407111926505, "centered_acc": -0.0011376564277588337, "centered_acc_ci": NaN, "centered_acc_runs": [-0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337], "ci": NaN}, "openbookqa": {"acc_norm": 0.284, "acc_norm_ci": NaN, "acc_norm_runs": [0.284, 0.284, 0.284, 0.284, 0.284], "acc": 0.154, "acc_ci": NaN, "acc_runs": [0.154, 0.154, 0.154, 0.154, 0.154], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020186703693570774, "centered_acc": 0.0453333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.48697711128650356, "acc_ci": NaN, "acc_runs": [0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014047718393997596, "centered_acc": -0.026045777426992878, "centered_acc_ci": NaN, "centered_acc_runs": [-0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5, "acc_ci": NaN, "acc_runs": [0.5, 0.5, 0.5, 0.5, 0.5], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.050251890762960605, "centered_acc": 0.0, "centered_acc_ci": NaN, "centered_acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 391991.009755762, "perplexity_ci": NaN, "perplexity_runs": [391991.009755762, 391991.009755762, 391991.009755762, 391991.009755762, 391991.009755762], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.191, "acc_norm_ci": NaN, "acc_norm_runs": [0.191, 0.191, 0.191, 0.191, 0.191], "acc": 0.054000000000000006, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.054, 0.054, 0.054, 0.054, 0.054], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012436787112179482, "centered_acc": -0.07866666666666666, "centered_acc_ci": NaN, "centered_acc_runs": [-0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666], "ci": NaN}}, "summary": {"mean_acc_norm": 0.30199618627155245, "mean_acc_norm_ci": NaN, "mean_acc": 0.28663186851370415, "mean_acc_ci": NaN, "mean_centered_acc": -0.01628594929031859, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.30199618627155245, 0.30199618627155245, 0.30199618627155245, 0.30199618627155245, 0.30199618627155245], "mean_acc_runs": [0.28663186851370415, 0.28663186851370415, 0.28663186851370415, 0.28663186851370415, 0.28663186851370415], "mean_centered_acc_runs": [-0.01628594929031859, -0.01628594929031859, -0.01628594929031859, -0.01628594929031859, -0.01628594929031859], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
| 2 |
{"run_dir": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152", "run": "baseline-fulle-lm-wsd-from-1152", "checkpoint_path": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152/checkpoint-1224", "checkpoint": "checkpoint-1224", "global_step": 1224, "tokens_seen": 1283457024, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "babi", "boolq", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2513443537143995, "acc_norm_ci": NaN, "acc_norm_runs": [0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995], "acc": 0.25761800438159727, "acc_ci": NaN, "acc_runs": [0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0043289955103121775, "centered_acc": 0.001792471619199339, "centered_acc_ci": NaN, "centered_acc_runs": [0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339], "ci": NaN}, "piqa": {"acc_norm": 0.5603917301414582, "acc_norm_ci": NaN, "acc_norm_runs": [0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582], "acc": 0.5680087051142546, "acc_ci": NaN, "acc_runs": [0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011580417248656388, "centered_acc": 0.12078346028291631, "centered_acc_ci": NaN, "centered_acc_runs": [0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631], "ci": NaN}, "arc_easy": {"acc_norm": 0.2760942760942761, "acc_norm_ci": NaN, "acc_norm_runs": [0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761], "acc": 0.2609427609427609, "acc_ci": NaN, "acc_runs": [0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00917355987383544, "centered_acc": 0.03479236812570149, "centered_acc_ci": NaN, "centered_acc_runs": [0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149], "ci": NaN}, "arc_challenge": {"acc_norm": 0.24914675767918087, "acc_norm_ci": NaN, "acc_norm_runs": [0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087], "acc": 0.20648464163822525, "acc_ci": NaN, "acc_runs": [0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012639407111926505, "centered_acc": -0.0011376564277588337, "centered_acc_ci": NaN, "centered_acc_runs": [-0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337], "ci": NaN}, "openbookqa": {"acc_norm": 0.284, "acc_norm_ci": NaN, "acc_norm_runs": [0.284, 0.284, 0.284, 0.284, 0.284], "acc": 0.154, "acc_ci": NaN, "acc_runs": [0.154, 0.154, 0.154, 0.154, 0.154], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020186703693570774, "centered_acc": 0.0453333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.48697711128650356, "acc_ci": NaN, "acc_runs": [0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014047718393997596, "centered_acc": -0.026045777426992878, "centered_acc_ci": NaN, "centered_acc_runs": [-0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5, "acc_ci": NaN, "acc_runs": [0.5, 0.5, 0.5, 0.5, 0.5], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.050251890762960605, "centered_acc": 0.0, "centered_acc_ci": NaN, "centered_acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 391991.009755762, "perplexity_ci": NaN, "perplexity_runs": [391991.009755762, 391991.009755762, 391991.009755762, 391991.009755762, 391991.009755762], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.191, "acc_norm_ci": NaN, "acc_norm_runs": [0.191, 0.191, 0.191, 0.191, 0.191], "acc": 0.054000000000000006, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.054, 0.054, 0.054, 0.054, 0.054], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012436787112179482, "centered_acc": -0.07866666666666666, "centered_acc_ci": NaN, "centered_acc_runs": [-0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666], "ci": NaN}, "babi": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": null, "acc_ci": 0.0, "acc_runs": [], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": 0.0}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}}, "summary": {"mean_acc_norm": 0.3019961862715526, "mean_acc_norm_ci": 0.04532134416790795, "mean_acc": 0.30849040885558277, "mean_acc_ci": 0.05121294953603617, "mean_centered_acc": -0.009242191906413072, "mean_centered_acc_ci": 0.02663212685397144, "mean_acc_norm_runs": [0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.284, 0.284, 0.284, 0.284, 0.284, 0.191, 0.191, 0.191, 0.191, 0.191], "mean_acc_runs": [0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.154, 0.154, 0.154, 0.154, 0.154, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.5, 0.5, 0.5, 0.5, 0.5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.054, 0.054, 0.054, 0.054, 0.054, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "mean_centered_acc_runs": [0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.0, 0.0, 0.0, 0.0, 0.0, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.04532134416790795}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152", "run": "baseline-fulle-lm-wsd-from-1152", "checkpoint_path": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152/checkpoint-1224", "checkpoint": "checkpoint-1224", "global_step": 1224, "tokens_seen": 1283457024, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2513443537143995, "acc_norm_ci": NaN, "acc_norm_runs": [0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995], "acc": 0.25761800438159727, "acc_ci": NaN, "acc_runs": [0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0043289955103121775, "centered_acc": 0.001792471619199339, "centered_acc_ci": NaN, "centered_acc_runs": [0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339], "ci": NaN}, "piqa": {"acc_norm": 0.5603917301414582, "acc_norm_ci": NaN, "acc_norm_runs": [0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582], "acc": 0.5680087051142546, "acc_ci": NaN, "acc_runs": [0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011580417248656388, "centered_acc": 0.12078346028291631, "centered_acc_ci": NaN, "centered_acc_runs": [0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631], "ci": NaN}, "arc_easy": {"acc_norm": 0.2760942760942761, "acc_norm_ci": NaN, "acc_norm_runs": [0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761], "acc": 0.2609427609427609, "acc_ci": NaN, "acc_runs": [0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00917355987383544, "centered_acc": 0.03479236812570149, "centered_acc_ci": NaN, "centered_acc_runs": [0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149], "ci": NaN}, "arc_challenge": {"acc_norm": 0.24914675767918087, "acc_norm_ci": NaN, "acc_norm_runs": [0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087], "acc": 0.20648464163822525, "acc_ci": NaN, "acc_runs": [0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012639407111926505, "centered_acc": -0.0011376564277588337, "centered_acc_ci": NaN, "centered_acc_runs": [-0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337], "ci": NaN}, "openbookqa": {"acc_norm": 0.284, "acc_norm_ci": NaN, "acc_norm_runs": [0.284, 0.284, 0.284, 0.284, 0.284], "acc": 0.154, "acc_ci": NaN, "acc_runs": [0.154, 0.154, 0.154, 0.154, 0.154], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020186703693570774, "centered_acc": 0.0453333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.48697711128650356, "acc_ci": NaN, "acc_runs": [0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014047718393997596, "centered_acc": -0.026045777426992878, "centered_acc_ci": NaN, "centered_acc_runs": [-0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5, "acc_ci": NaN, "acc_runs": [0.5, 0.5, 0.5, 0.5, 0.5], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.050251890762960605, "centered_acc": 0.0, "centered_acc_ci": NaN, "centered_acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 391991.009755762, "perplexity_ci": NaN, "perplexity_runs": [391991.009755762, 391991.009755762, 391991.009755762, 391991.009755762, 391991.009755762], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.191, "acc_norm_ci": NaN, "acc_norm_runs": [0.191, 0.191, 0.191, 0.191, 0.191], "acc": 0.054000000000000006, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.054, 0.054, 0.054, 0.054, 0.054], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012436787112179482, "centered_acc": -0.07866666666666666, "centered_acc_ci": NaN, "centered_acc_runs": [-0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666], "ci": NaN}}, "summary": {"mean_acc_norm": 0.30199618627155245, "mean_acc_norm_ci": NaN, "mean_acc": 0.28663186851370415, "mean_acc_ci": NaN, "mean_centered_acc": -0.01628594929031859, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.30199618627155245, 0.30199618627155245, 0.30199618627155245, 0.30199618627155245, 0.30199618627155245], "mean_acc_runs": [0.28663186851370415, 0.28663186851370415, 0.28663186851370415, 0.28663186851370415, 0.28663186851370415], "mean_centered_acc_runs": [-0.01628594929031859, -0.01628594929031859, -0.01628594929031859, -0.01628594929031859, -0.01628594929031859], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
| 2 |
{"run_dir": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152", "run": "baseline-fulle-lm-wsd-from-1152", "checkpoint_path": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152/checkpoint-1224", "checkpoint": "checkpoint-1224", "global_step": 1224, "tokens_seen": 1283457024, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "babi", "boolq", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2513443537143995, "acc_norm_ci": NaN, "acc_norm_runs": [0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995], "acc": 0.25761800438159727, "acc_ci": NaN, "acc_runs": [0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0043289955103121775, "centered_acc": 0.001792471619199339, "centered_acc_ci": NaN, "centered_acc_runs": [0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339], "ci": NaN}, "piqa": {"acc_norm": 0.5603917301414582, "acc_norm_ci": NaN, "acc_norm_runs": [0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582], "acc": 0.5680087051142546, "acc_ci": NaN, "acc_runs": [0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011580417248656388, "centered_acc": 0.12078346028291631, "centered_acc_ci": NaN, "centered_acc_runs": [0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631], "ci": NaN}, "arc_easy": {"acc_norm": 0.2760942760942761, "acc_norm_ci": NaN, "acc_norm_runs": [0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761], "acc": 0.2609427609427609, "acc_ci": NaN, "acc_runs": [0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00917355987383544, "centered_acc": 0.03479236812570149, "centered_acc_ci": NaN, "centered_acc_runs": [0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149], "ci": NaN}, "arc_challenge": {"acc_norm": 0.24914675767918087, "acc_norm_ci": NaN, "acc_norm_runs": [0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087], "acc": 0.20648464163822525, "acc_ci": NaN, "acc_runs": [0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012639407111926505, "centered_acc": -0.0011376564277588337, "centered_acc_ci": NaN, "centered_acc_runs": [-0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337], "ci": NaN}, "openbookqa": {"acc_norm": 0.284, "acc_norm_ci": NaN, "acc_norm_runs": [0.284, 0.284, 0.284, 0.284, 0.284], "acc": 0.154, "acc_ci": NaN, "acc_runs": [0.154, 0.154, 0.154, 0.154, 0.154], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020186703693570774, "centered_acc": 0.0453333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.48697711128650356, "acc_ci": NaN, "acc_runs": [0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014047718393997596, "centered_acc": -0.026045777426992878, "centered_acc_ci": NaN, "centered_acc_runs": [-0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5, "acc_ci": NaN, "acc_runs": [0.5, 0.5, 0.5, 0.5, 0.5], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.050251890762960605, "centered_acc": 0.0, "centered_acc_ci": NaN, "centered_acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 391991.009755762, "perplexity_ci": NaN, "perplexity_runs": [391991.009755762, 391991.009755762, 391991.009755762, 391991.009755762, 391991.009755762], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.191, "acc_norm_ci": NaN, "acc_norm_runs": [0.191, 0.191, 0.191, 0.191, 0.191], "acc": 0.054000000000000006, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.054, 0.054, 0.054, 0.054, 0.054], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012436787112179482, "centered_acc": -0.07866666666666666, "centered_acc_ci": NaN, "centered_acc_runs": [-0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666], "ci": NaN}, "babi": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": null, "acc_ci": 0.0, "acc_runs": [], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": 0.0}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}}, "summary": {"mean_acc_norm": 0.3019961862715526, "mean_acc_norm_ci": 0.04532134416790795, "mean_acc": 0.30849040885558277, "mean_acc_ci": 0.05121294953603617, "mean_centered_acc": -0.009242191906413072, "mean_centered_acc_ci": 0.02663212685397144, "mean_acc_norm_runs": [0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.284, 0.284, 0.284, 0.284, 0.284, 0.191, 0.191, 0.191, 0.191, 0.191], "mean_acc_runs": [0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.154, 0.154, 0.154, 0.154, 0.154, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.5, 0.5, 0.5, 0.5, 0.5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.054, 0.054, 0.054, 0.054, 0.054, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "mean_centered_acc_runs": [0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.0, 0.0, 0.0, 0.0, 0.0, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.04532134416790795}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
| 3 |
+
{"run_dir": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152", "run": "baseline-fulle-lm-wsd-from-1152", "checkpoint_path": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152/checkpoint-1224", "checkpoint": "checkpoint-1224", "global_step": 1224, "tokens_seen": 1283457024, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "babi", "boolq", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2513443537143995, "acc_norm_ci": NaN, "acc_norm_runs": [0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995], "acc": 0.25761800438159727, "acc_ci": NaN, "acc_runs": [0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0043289955103121775, "centered_acc": 0.001792471619199339, "centered_acc_ci": NaN, "centered_acc_runs": [0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339], "ci": NaN}, "piqa": {"acc_norm": 0.5603917301414582, "acc_norm_ci": NaN, "acc_norm_runs": [0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582], "acc": 0.5680087051142546, "acc_ci": NaN, "acc_runs": [0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011580417248656388, "centered_acc": 0.12078346028291631, "centered_acc_ci": NaN, "centered_acc_runs": [0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631], "ci": NaN}, "arc_easy": {"acc_norm": 0.2760942760942761, "acc_norm_ci": NaN, "acc_norm_runs": [0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761], "acc": 0.2609427609427609, "acc_ci": NaN, "acc_runs": [0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00917355987383544, "centered_acc": 0.03479236812570149, "centered_acc_ci": NaN, "centered_acc_runs": [0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149], "ci": NaN}, "arc_challenge": {"acc_norm": 0.24914675767918087, "acc_norm_ci": NaN, "acc_norm_runs": [0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087], "acc": 0.20648464163822525, "acc_ci": NaN, "acc_runs": [0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012639407111926505, "centered_acc": -0.0011376564277588337, "centered_acc_ci": NaN, "centered_acc_runs": [-0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337], "ci": NaN}, "openbookqa": {"acc_norm": 0.284, "acc_norm_ci": NaN, "acc_norm_runs": [0.284, 0.284, 0.284, 0.284, 0.284], "acc": 0.154, "acc_ci": NaN, "acc_runs": [0.154, 0.154, 0.154, 0.154, 0.154], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020186703693570774, "centered_acc": 0.0453333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.48697711128650356, "acc_ci": NaN, "acc_runs": [0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014047718393997596, "centered_acc": -0.026045777426992878, "centered_acc_ci": NaN, "centered_acc_runs": [-0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5, "acc_ci": NaN, "acc_runs": [0.5, 0.5, 0.5, 0.5, 0.5], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.050251890762960605, "centered_acc": 0.0, "centered_acc_ci": NaN, "centered_acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 391991.009755762, "perplexity_ci": NaN, "perplexity_runs": [391991.009755762, 391991.009755762, 391991.009755762, 391991.009755762, 391991.009755762], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.191, "acc_norm_ci": NaN, "acc_norm_runs": [0.191, 0.191, 0.191, 0.191, 0.191], "acc": 0.054000000000000006, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.054, 0.054, 0.054, 0.054, 0.054], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012436787112179482, "centered_acc": -0.07866666666666666, "centered_acc_ci": NaN, "centered_acc_runs": [-0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "babi": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": null, "acc_ci": 0.0, "acc_runs": [], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": 0.0}}, "summary": {"mean_acc_norm": 0.3019961862715526, "mean_acc_norm_ci": 0.04532134416790795, "mean_acc": 0.30849040885558277, "mean_acc_ci": 0.05121294953603617, "mean_centered_acc": -0.009242191906413072, "mean_centered_acc_ci": 0.02663212685397144, "mean_acc_norm_runs": [0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.284, 0.284, 0.284, 0.284, 0.284, 0.191, 0.191, 0.191, 0.191, 0.191], "mean_acc_runs": [0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.154, 0.154, 0.154, 0.154, 0.154, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.5, 0.5, 0.5, 0.5, 0.5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.054, 0.054, 0.054, 0.054, 0.054, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "mean_centered_acc_runs": [0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.0, 0.0, 0.0, 0.0, 0.0, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.04532134416790795}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
| 4 |
+
{"run_dir": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152", "run": "baseline-fulle-lm-wsd-from-1152", "checkpoint_path": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152/checkpoint-1224", "checkpoint": "checkpoint-1224", "global_step": 1224, "tokens_seen": 1283457024, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq", "rte", "commonsense_qa"], "tasks": {"hellaswag": {"acc_norm": 0.2513443537143995, "acc_norm_ci": NaN, "acc_norm_runs": [0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995], "acc": 0.25761800438159727, "acc_ci": NaN, "acc_runs": [0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0043289955103121775, "centered_acc": 0.001792471619199339, "centered_acc_ci": NaN, "centered_acc_runs": [0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339], "ci": NaN}, "piqa": {"acc_norm": 0.5603917301414582, "acc_norm_ci": NaN, "acc_norm_runs": [0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582], "acc": 0.5680087051142546, "acc_ci": NaN, "acc_runs": [0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011580417248656388, "centered_acc": 0.12078346028291631, "centered_acc_ci": NaN, "centered_acc_runs": [0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631], "ci": NaN}, "arc_easy": {"acc_norm": 0.2760942760942761, "acc_norm_ci": NaN, "acc_norm_runs": [0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761], "acc": 0.2609427609427609, "acc_ci": NaN, "acc_runs": [0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00917355987383544, "centered_acc": 0.03479236812570149, "centered_acc_ci": NaN, "centered_acc_runs": [0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149], "ci": NaN}, "arc_challenge": {"acc_norm": 0.24914675767918087, "acc_norm_ci": NaN, "acc_norm_runs": [0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087], "acc": 0.20648464163822525, "acc_ci": NaN, "acc_runs": [0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012639407111926505, "centered_acc": -0.0011376564277588337, "centered_acc_ci": NaN, "centered_acc_runs": [-0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337], "ci": NaN}, "openbookqa": {"acc_norm": 0.284, "acc_norm_ci": NaN, "acc_norm_runs": [0.284, 0.284, 0.284, 0.284, 0.284], "acc": 0.154, "acc_ci": NaN, "acc_runs": [0.154, 0.154, 0.154, 0.154, 0.154], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020186703693570774, "centered_acc": 0.0453333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.48697711128650356, "acc_ci": NaN, "acc_runs": [0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014047718393997596, "centered_acc": -0.026045777426992878, "centered_acc_ci": NaN, "centered_acc_runs": [-0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5, "acc_ci": NaN, "acc_runs": [0.5, 0.5, 0.5, 0.5, 0.5], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.050251890762960605, "centered_acc": 0.0, "centered_acc_ci": NaN, "centered_acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 391991.60405657435, "perplexity_ci": NaN, "perplexity_runs": [391991.60405657435, 391991.60405657435, 391991.60405657435, 391991.60405657435, 391991.60405657435], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.191, "acc_norm_ci": NaN, "acc_norm_runs": [0.191, 0.191, 0.191, 0.191, 0.191], "acc": 0.054000000000000006, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.054, 0.054, 0.054, 0.054, 0.054], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012436787112179482, "centered_acc": -0.07866666666666666, "centered_acc_ci": NaN, "centered_acc_runs": [-0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.30199618627155245, "mean_acc_norm_ci": NaN, "mean_acc": 0.29909464109605044, "mean_acc_ci": NaN, "mean_centered_acc": -0.009242191906413075, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.30199618627155245, 0.30199618627155245, 0.30199618627155245, 0.30199618627155245, 0.30199618627155245], "mean_acc_runs": [0.29909464109605044, 0.29909464109605044, 0.29909464109605044, 0.29909464109605044, 0.29909464109605044], "mean_centered_acc_runs": [-0.009242191906413075, -0.009242191906413075, -0.009242191906413075, -0.009242191906413075, -0.009242191906413075], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
| 5 |
+
{"run_dir": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152", "run": "baseline-fulle-lm-wsd-from-1152", "checkpoint_path": "/workspace/concept_model/runs/baseline-fulle-lm-wsd-from-1152/checkpoint-1224", "checkpoint": "checkpoint-1224", "global_step": 1224, "tokens_seen": 1283457024, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "blimp", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2513443537143995, "acc_norm_ci": NaN, "acc_norm_runs": [0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995], "acc": 0.25761800438159727, "acc_ci": NaN, "acc_runs": [0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0043289955103121775, "centered_acc": 0.001792471619199339, "centered_acc_ci": NaN, "centered_acc_runs": [0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339], "ci": NaN}, "piqa": {"acc_norm": 0.5603917301414582, "acc_norm_ci": NaN, "acc_norm_runs": [0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582], "acc": 0.5680087051142546, "acc_ci": NaN, "acc_runs": [0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011580417248656388, "centered_acc": 0.12078346028291631, "centered_acc_ci": NaN, "centered_acc_runs": [0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631], "ci": NaN}, "arc_easy": {"acc_norm": 0.2760942760942761, "acc_norm_ci": NaN, "acc_norm_runs": [0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761], "acc": 0.2609427609427609, "acc_ci": NaN, "acc_runs": [0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00917355987383544, "centered_acc": 0.03479236812570149, "centered_acc_ci": NaN, "centered_acc_runs": [0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149], "ci": NaN}, "arc_challenge": {"acc_norm": 0.24914675767918087, "acc_norm_ci": NaN, "acc_norm_runs": [0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087], "acc": 0.20648464163822525, "acc_ci": NaN, "acc_runs": [0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012639407111926505, "centered_acc": -0.0011376564277588337, "centered_acc_ci": NaN, "centered_acc_runs": [-0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337], "ci": NaN}, "openbookqa": {"acc_norm": 0.284, "acc_norm_ci": NaN, "acc_norm_runs": [0.284, 0.284, 0.284, 0.284, 0.284], "acc": 0.154, "acc_ci": NaN, "acc_runs": [0.154, 0.154, 0.154, 0.154, 0.154], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020186703693570774, "centered_acc": 0.0453333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.48697711128650356, "acc_ci": NaN, "acc_runs": [0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014047718393997596, "centered_acc": -0.026045777426992878, "centered_acc_ci": NaN, "centered_acc_runs": [-0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5, "acc_ci": NaN, "acc_runs": [0.5, 0.5, 0.5, 0.5, 0.5], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.050251890762960605, "centered_acc": 0.0, "centered_acc_ci": NaN, "centered_acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 391991.60405657435, "perplexity_ci": NaN, "perplexity_runs": [391991.60405657435, 391991.60405657435, 391991.60405657435, 391991.60405657435, 391991.60405657435], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.191, "acc_norm_ci": NaN, "acc_norm_runs": [0.191, 0.191, 0.191, 0.191, 0.191], "acc": 0.054000000000000006, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.054, 0.054, 0.054, 0.054, 0.054], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012436787112179482, "centered_acc": -0.07866666666666666, "centered_acc_ci": NaN, "centered_acc_runs": [-0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "blimp": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.6510597014925373, "acc_ci": NaN, "acc_runs": [0.6510597014925373, 0.6510597014925373, 0.6510597014925373, 0.6510597014925373, 0.6510597014925373], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0016767374089327464, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.3019961862715526, "mean_acc_norm_ci": 0.04532134416790795, "mean_acc": 0.3261688765111649, "mean_acc_ci": 0.04963449075055859, "mean_centered_acc": -0.009242191906413072, "mean_centered_acc_ci": 0.02663212685397144, "mean_acc_norm_runs": [0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.2513443537143995, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.5603917301414582, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.2760942760942761, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.24914675767918087, 0.284, 0.284, 0.284, 0.284, 0.284, 0.191, 0.191, 0.191, 0.191, 0.191], "mean_acc_runs": [0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.25761800438159727, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.5680087051142546, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.2609427609427609, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.20648464163822525, 0.154, 0.154, 0.154, 0.154, 0.154, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.48697711128650356, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.5, 0.5, 0.5, 0.5, 0.5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.054, 0.054, 0.054, 0.054, 0.054, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.6510597014925373, 0.6510597014925373, 0.6510597014925373, 0.6510597014925373, 0.6510597014925373], "mean_centered_acc_runs": [0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.001792471619199339, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.12078346028291631, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, 0.03479236812570149, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, -0.0011376564277588337, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, 0.0453333333333333, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.026045777426992878, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.0, 0.0, 0.0, 0.0, 0.0, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, -0.07866666666666666, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.04532134416790795}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
results/i5-exp-fulle-lm-wsd-from-4320/zero_shot_eval/checkpoint-4464.json
CHANGED
|
@@ -9,16 +9,18 @@
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
| 12 |
-
"hellaswag",
|
| 13 |
-
"piqa",
|
| 14 |
-
"arc_easy",
|
| 15 |
"arc_challenge",
|
| 16 |
-
"
|
| 17 |
-
"winogrande",
|
| 18 |
"boolq",
|
|
|
|
| 19 |
"copa",
|
|
|
|
| 20 |
"lambada_openai",
|
| 21 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
],
|
| 23 |
"tasks": {
|
| 24 |
"hellaswag": {
|
|
@@ -336,37 +338,217 @@
|
|
| 336 |
-0.08933333333333333
|
| 337 |
],
|
| 338 |
"ci": NaN
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 339 |
}
|
| 340 |
},
|
| 341 |
"summary": {
|
| 342 |
-
"mean_acc_norm": 0.
|
| 343 |
-
"mean_acc_norm_ci":
|
| 344 |
-
"mean_acc": 0.
|
| 345 |
-
"mean_acc_ci":
|
| 346 |
-
"mean_centered_acc": 0.
|
| 347 |
-
"mean_centered_acc_ci":
|
| 348 |
"mean_acc_norm_runs": [
|
| 349 |
-
0.
|
| 350 |
-
0.
|
| 351 |
-
0.
|
| 352 |
-
0.
|
| 353 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 354 |
],
|
| 355 |
"mean_acc_runs": [
|
| 356 |
-
0.
|
| 357 |
-
0.
|
| 358 |
-
0.
|
| 359 |
-
0.
|
| 360 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 361 |
],
|
| 362 |
"mean_centered_acc_runs": [
|
| 363 |
-
0.
|
| 364 |
-
0.
|
| 365 |
-
0.
|
| 366 |
-
0.
|
| 367 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 368 |
],
|
| 369 |
-
"ci":
|
| 370 |
},
|
| 371 |
"batch_size_per_device": 8,
|
| 372 |
"device": "cuda:0",
|
|
@@ -376,4 +558,4 @@
|
|
| 376 |
"tokenizer_has_l2r_pred": true,
|
| 377 |
"tokenizer_has_next_1_pred": true,
|
| 378 |
"max_next_i_train": 5
|
| 379 |
-
}
|
|
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
|
|
|
|
|
|
|
|
|
| 12 |
"arc_challenge",
|
| 13 |
+
"arc_easy",
|
|
|
|
| 14 |
"boolq",
|
| 15 |
+
"commonsense_qa",
|
| 16 |
"copa",
|
| 17 |
+
"hellaswag",
|
| 18 |
"lambada_openai",
|
| 19 |
+
"openbookqa",
|
| 20 |
+
"piqa",
|
| 21 |
+
"rte",
|
| 22 |
+
"sciq",
|
| 23 |
+
"winogrande"
|
| 24 |
],
|
| 25 |
"tasks": {
|
| 26 |
"hellaswag": {
|
|
|
|
| 338 |
-0.08933333333333333
|
| 339 |
],
|
| 340 |
"ci": NaN
|
| 341 |
+
},
|
| 342 |
+
"rte": {
|
| 343 |
+
"acc_norm": null,
|
| 344 |
+
"acc_norm_ci": 0.0,
|
| 345 |
+
"acc_norm_runs": [],
|
| 346 |
+
"acc": 0.5270758122743683,
|
| 347 |
+
"acc_ci": NaN,
|
| 348 |
+
"acc_runs": [
|
| 349 |
+
0.5270758122743683,
|
| 350 |
+
0.5270758122743683,
|
| 351 |
+
0.5270758122743683,
|
| 352 |
+
0.5270758122743683,
|
| 353 |
+
0.5270758122743683
|
| 354 |
+
],
|
| 355 |
+
"perplexity": null,
|
| 356 |
+
"perplexity_ci": 0.0,
|
| 357 |
+
"perplexity_runs": [],
|
| 358 |
+
"stderr": 0.030052303463143738,
|
| 359 |
+
"centered_acc": 0.054151624548736566,
|
| 360 |
+
"centered_acc_ci": NaN,
|
| 361 |
+
"centered_acc_runs": [
|
| 362 |
+
0.054151624548736566,
|
| 363 |
+
0.054151624548736566,
|
| 364 |
+
0.054151624548736566,
|
| 365 |
+
0.054151624548736566,
|
| 366 |
+
0.054151624548736566
|
| 367 |
+
],
|
| 368 |
+
"ci": NaN
|
| 369 |
+
},
|
| 370 |
+
"commonsense_qa": {
|
| 371 |
+
"acc_norm": null,
|
| 372 |
+
"acc_norm_ci": 0.0,
|
| 373 |
+
"acc_norm_runs": [],
|
| 374 |
+
"acc": 0.19574119574119575,
|
| 375 |
+
"acc_ci": NaN,
|
| 376 |
+
"acc_runs": [
|
| 377 |
+
0.19574119574119575,
|
| 378 |
+
0.19574119574119575,
|
| 379 |
+
0.19574119574119575,
|
| 380 |
+
0.19574119574119575,
|
| 381 |
+
0.19574119574119575
|
| 382 |
+
],
|
| 383 |
+
"perplexity": null,
|
| 384 |
+
"perplexity_ci": 0.0,
|
| 385 |
+
"perplexity_runs": [],
|
| 386 |
+
"stderr": 0.011359497363584397,
|
| 387 |
+
"centered_acc": null,
|
| 388 |
+
"centered_acc_ci": 0.0,
|
| 389 |
+
"centered_acc_runs": [],
|
| 390 |
+
"ci": NaN
|
| 391 |
}
|
| 392 |
},
|
| 393 |
"summary": {
|
| 394 |
+
"mean_acc_norm": 0.30026004206925716,
|
| 395 |
+
"mean_acc_norm_ci": 0.042481836721542376,
|
| 396 |
+
"mean_acc": 0.3298487599795072,
|
| 397 |
+
"mean_acc_ci": 0.04970978100318946,
|
| 398 |
+
"mean_centered_acc": 0.014511603300729349,
|
| 399 |
+
"mean_centered_acc_ci": 0.031496150671264184,
|
| 400 |
"mean_acc_norm_runs": [
|
| 401 |
+
0.2544313881696873,
|
| 402 |
+
0.2544313881696873,
|
| 403 |
+
0.2544313881696873,
|
| 404 |
+
0.2544313881696873,
|
| 405 |
+
0.2544313881696873,
|
| 406 |
+
0.5380848748639826,
|
| 407 |
+
0.5380848748639826,
|
| 408 |
+
0.5380848748639826,
|
| 409 |
+
0.5380848748639826,
|
| 410 |
+
0.5380848748639826,
|
| 411 |
+
0.2777777777777778,
|
| 412 |
+
0.2777777777777778,
|
| 413 |
+
0.2777777777777778,
|
| 414 |
+
0.2777777777777778,
|
| 415 |
+
0.2777777777777778,
|
| 416 |
+
0.25426621160409557,
|
| 417 |
+
0.25426621160409557,
|
| 418 |
+
0.25426621160409557,
|
| 419 |
+
0.25426621160409557,
|
| 420 |
+
0.25426621160409557,
|
| 421 |
+
0.294,
|
| 422 |
+
0.294,
|
| 423 |
+
0.294,
|
| 424 |
+
0.294,
|
| 425 |
+
0.294,
|
| 426 |
+
0.183,
|
| 427 |
+
0.183,
|
| 428 |
+
0.183,
|
| 429 |
+
0.183,
|
| 430 |
+
0.183
|
| 431 |
],
|
| 432 |
"mean_acc_runs": [
|
| 433 |
+
0.2560246962756423,
|
| 434 |
+
0.2560246962756423,
|
| 435 |
+
0.2560246962756423,
|
| 436 |
+
0.2560246962756423,
|
| 437 |
+
0.2560246962756423,
|
| 438 |
+
0.5457018498367792,
|
| 439 |
+
0.5457018498367792,
|
| 440 |
+
0.5457018498367792,
|
| 441 |
+
0.5457018498367792,
|
| 442 |
+
0.5457018498367792,
|
| 443 |
+
0.26304713804713803,
|
| 444 |
+
0.26304713804713803,
|
| 445 |
+
0.26304713804713803,
|
| 446 |
+
0.26304713804713803,
|
| 447 |
+
0.26304713804713803,
|
| 448 |
+
0.2090443686006826,
|
| 449 |
+
0.2090443686006826,
|
| 450 |
+
0.2090443686006826,
|
| 451 |
+
0.2090443686006826,
|
| 452 |
+
0.2090443686006826,
|
| 453 |
+
0.164,
|
| 454 |
+
0.164,
|
| 455 |
+
0.164,
|
| 456 |
+
0.164,
|
| 457 |
+
0.164,
|
| 458 |
+
0.5201262825572218,
|
| 459 |
+
0.5201262825572218,
|
| 460 |
+
0.5201262825572218,
|
| 461 |
+
0.5201262825572218,
|
| 462 |
+
0.5201262825572218,
|
| 463 |
+
0.3782874617737003,
|
| 464 |
+
0.3782874617737003,
|
| 465 |
+
0.3782874617737003,
|
| 466 |
+
0.3782874617737003,
|
| 467 |
+
0.3782874617737003,
|
| 468 |
+
0.6,
|
| 469 |
+
0.6,
|
| 470 |
+
0.6,
|
| 471 |
+
0.6,
|
| 472 |
+
0.6,
|
| 473 |
+
0.0,
|
| 474 |
+
0.0,
|
| 475 |
+
0.0,
|
| 476 |
+
0.0,
|
| 477 |
+
0.0,
|
| 478 |
+
0.051,
|
| 479 |
+
0.051,
|
| 480 |
+
0.051,
|
| 481 |
+
0.051,
|
| 482 |
+
0.051,
|
| 483 |
+
0.5270758122743683,
|
| 484 |
+
0.5270758122743683,
|
| 485 |
+
0.5270758122743683,
|
| 486 |
+
0.5270758122743683,
|
| 487 |
+
0.5270758122743683,
|
| 488 |
+
0.19574119574119575,
|
| 489 |
+
0.19574119574119575,
|
| 490 |
+
0.19574119574119575,
|
| 491 |
+
0.19574119574119575,
|
| 492 |
+
0.19574119574119575,
|
| 493 |
+
0.5779850746268657,
|
| 494 |
+
0.5779850746268657,
|
| 495 |
+
0.5779850746268657,
|
| 496 |
+
0.5779850746268657,
|
| 497 |
+
0.5779850746268657
|
| 498 |
],
|
| 499 |
"mean_centered_acc_runs": [
|
| 500 |
+
0.005908517559583067,
|
| 501 |
+
0.005908517559583067,
|
| 502 |
+
0.005908517559583067,
|
| 503 |
+
0.005908517559583067,
|
| 504 |
+
0.005908517559583067,
|
| 505 |
+
0.0761697497279652,
|
| 506 |
+
0.0761697497279652,
|
| 507 |
+
0.0761697497279652,
|
| 508 |
+
0.0761697497279652,
|
| 509 |
+
0.0761697497279652,
|
| 510 |
+
0.037037037037037056,
|
| 511 |
+
0.037037037037037056,
|
| 512 |
+
0.037037037037037056,
|
| 513 |
+
0.037037037037037056,
|
| 514 |
+
0.037037037037037056,
|
| 515 |
+
0.0056882821387940945,
|
| 516 |
+
0.0056882821387940945,
|
| 517 |
+
0.0056882821387940945,
|
| 518 |
+
0.0056882821387940945,
|
| 519 |
+
0.0056882821387940945,
|
| 520 |
+
0.058666666666666645,
|
| 521 |
+
0.058666666666666645,
|
| 522 |
+
0.058666666666666645,
|
| 523 |
+
0.058666666666666645,
|
| 524 |
+
0.058666666666666645,
|
| 525 |
+
0.04025256511444364,
|
| 526 |
+
0.04025256511444364,
|
| 527 |
+
0.04025256511444364,
|
| 528 |
+
0.04025256511444364,
|
| 529 |
+
0.04025256511444364,
|
| 530 |
+
-0.24342507645259936,
|
| 531 |
+
-0.24342507645259936,
|
| 532 |
+
-0.24342507645259936,
|
| 533 |
+
-0.24342507645259936,
|
| 534 |
+
-0.24342507645259936,
|
| 535 |
+
0.19999999999999996,
|
| 536 |
+
0.19999999999999996,
|
| 537 |
+
0.19999999999999996,
|
| 538 |
+
0.19999999999999996,
|
| 539 |
+
0.19999999999999996,
|
| 540 |
+
-0.08933333333333333,
|
| 541 |
+
-0.08933333333333333,
|
| 542 |
+
-0.08933333333333333,
|
| 543 |
+
-0.08933333333333333,
|
| 544 |
+
-0.08933333333333333,
|
| 545 |
+
0.054151624548736566,
|
| 546 |
+
0.054151624548736566,
|
| 547 |
+
0.054151624548736566,
|
| 548 |
+
0.054151624548736566,
|
| 549 |
+
0.054151624548736566
|
| 550 |
],
|
| 551 |
+
"ci": 0.042481836721542376
|
| 552 |
},
|
| 553 |
"batch_size_per_device": 8,
|
| 554 |
"device": "cuda:0",
|
|
|
|
| 558 |
"tokenizer_has_l2r_pred": true,
|
| 559 |
"tokenizer_has_next_1_pred": true,
|
| 560 |
"max_next_i_train": 5
|
| 561 |
+
}
|
results/i5-exp-fulle-lm-wsd-from-4320/zero_shot_eval/summary.jsonl
CHANGED
|
@@ -1 +1,3 @@
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/i5-exp-fulle-lm-wsd-from-4320", "run": "i5-exp-fulle-lm-wsd-from-4320", "checkpoint_path": "/workspace/concept_model/runs/i5-exp-fulle-lm-wsd-from-4320/checkpoint-4464", "checkpoint": "checkpoint-4464", "global_step": 4464, "tokens_seen": 4680843264, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2544313881696873, "acc_norm_ci": NaN, "acc_norm_runs": [0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873], "acc": 0.2560246962756423, "acc_ci": NaN, "acc_runs": [0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00434650985067984, "centered_acc": 0.005908517559583067, "centered_acc_ci": NaN, "centered_acc_runs": [0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067], "ci": NaN}, "piqa": {"acc_norm": 0.5380848748639826, "acc_norm_ci": NaN, "acc_norm_runs": [0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826], "acc": 0.5457018498367792, "acc_ci": NaN, "acc_runs": [0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011631933367846934, "centered_acc": 0.0761697497279652, "centered_acc_ci": NaN, "centered_acc_runs": [0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652], "ci": NaN}, "arc_easy": {"acc_norm": 0.2777777777777778, "acc_norm_ci": NaN, "acc_norm_runs": [0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778], "acc": 0.26304713804713803, "acc_ci": NaN, "acc_runs": [0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009190779909650122, "centered_acc": 0.037037037037037056, "centered_acc_ci": NaN, "centered_acc_runs": [0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056], "ci": NaN}, "arc_challenge": {"acc_norm": 0.25426621160409557, "acc_norm_ci": NaN, "acc_norm_runs": [0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557], "acc": 0.2090443686006826, "acc_ci": NaN, "acc_runs": [0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012724999945157845, "centered_acc": 0.0056882821387940945, "centered_acc_ci": NaN, "centered_acc_runs": [0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945], "ci": NaN}, "openbookqa": {"acc_norm": 0.294, "acc_norm_ci": NaN, "acc_norm_runs": [0.294, 0.294, 0.294, 0.294, 0.294], "acc": 0.164, "acc_ci": NaN, "acc_runs": [0.164, 0.164, 0.164, 0.164, 0.164], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.02039509548493655, "centered_acc": 0.058666666666666645, "centered_acc_ci": NaN, "centered_acc_runs": [0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5201262825572218, "acc_ci": NaN, "acc_runs": [0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.01404109666434435, "centered_acc": 0.04025256511444364, "centered_acc_ci": NaN, "centered_acc_runs": [0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.6, "acc_ci": NaN, "acc_runs": [0.6, 0.6, 0.6, 0.6, 0.6], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0492365963917331, "centered_acc": 0.19999999999999996, "centered_acc_ci": NaN, "centered_acc_runs": [0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 11127748.30585826, "perplexity_ci": NaN, "perplexity_runs": [11127748.30585826, 11127748.30585826, 11127748.30585826, 11127748.30585826, 11127748.30585826], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.183, "acc_norm_ci": NaN, "acc_norm_runs": [0.183, 0.183, 0.183, 0.183, 0.183], "acc": 0.051000000000000004, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.051, 0.051, 0.051, 0.051, 0.051], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012233587399477745, "centered_acc": -0.08933333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333], "ci": NaN}}, "summary": {"mean_acc_norm": 0.3002600420692572, "mean_acc_norm_ci": NaN, "mean_acc": 0.29872317970911644, "mean_acc_ci": NaN, "mean_centered_acc": 0.010107156495395217, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.3002600420692572, 0.3002600420692572, 0.3002600420692572, 0.3002600420692572, 0.3002600420692572], "mean_acc_runs": [0.29872317970911644, 0.29872317970911644, 0.29872317970911644, 0.29872317970911644, 0.29872317970911644], "mean_centered_acc_runs": [0.010107156495395217, 0.010107156495395217, 0.010107156495395217, 0.010107156495395217, 0.010107156495395217], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
|
|
|
|
|
|
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/i5-exp-fulle-lm-wsd-from-4320", "run": "i5-exp-fulle-lm-wsd-from-4320", "checkpoint_path": "/workspace/concept_model/runs/i5-exp-fulle-lm-wsd-from-4320/checkpoint-4464", "checkpoint": "checkpoint-4464", "global_step": 4464, "tokens_seen": 4680843264, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2544313881696873, "acc_norm_ci": NaN, "acc_norm_runs": [0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873], "acc": 0.2560246962756423, "acc_ci": NaN, "acc_runs": [0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00434650985067984, "centered_acc": 0.005908517559583067, "centered_acc_ci": NaN, "centered_acc_runs": [0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067], "ci": NaN}, "piqa": {"acc_norm": 0.5380848748639826, "acc_norm_ci": NaN, "acc_norm_runs": [0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826], "acc": 0.5457018498367792, "acc_ci": NaN, "acc_runs": [0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011631933367846934, "centered_acc": 0.0761697497279652, "centered_acc_ci": NaN, "centered_acc_runs": [0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652], "ci": NaN}, "arc_easy": {"acc_norm": 0.2777777777777778, "acc_norm_ci": NaN, "acc_norm_runs": [0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778], "acc": 0.26304713804713803, "acc_ci": NaN, "acc_runs": [0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009190779909650122, "centered_acc": 0.037037037037037056, "centered_acc_ci": NaN, "centered_acc_runs": [0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056], "ci": NaN}, "arc_challenge": {"acc_norm": 0.25426621160409557, "acc_norm_ci": NaN, "acc_norm_runs": [0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557], "acc": 0.2090443686006826, "acc_ci": NaN, "acc_runs": [0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012724999945157845, "centered_acc": 0.0056882821387940945, "centered_acc_ci": NaN, "centered_acc_runs": [0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945], "ci": NaN}, "openbookqa": {"acc_norm": 0.294, "acc_norm_ci": NaN, "acc_norm_runs": [0.294, 0.294, 0.294, 0.294, 0.294], "acc": 0.164, "acc_ci": NaN, "acc_runs": [0.164, 0.164, 0.164, 0.164, 0.164], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.02039509548493655, "centered_acc": 0.058666666666666645, "centered_acc_ci": NaN, "centered_acc_runs": [0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5201262825572218, "acc_ci": NaN, "acc_runs": [0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.01404109666434435, "centered_acc": 0.04025256511444364, "centered_acc_ci": NaN, "centered_acc_runs": [0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.6, "acc_ci": NaN, "acc_runs": [0.6, 0.6, 0.6, 0.6, 0.6], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0492365963917331, "centered_acc": 0.19999999999999996, "centered_acc_ci": NaN, "centered_acc_runs": [0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 11127748.30585826, "perplexity_ci": NaN, "perplexity_runs": [11127748.30585826, 11127748.30585826, 11127748.30585826, 11127748.30585826, 11127748.30585826], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.183, "acc_norm_ci": NaN, "acc_norm_runs": [0.183, 0.183, 0.183, 0.183, 0.183], "acc": 0.051000000000000004, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.051, 0.051, 0.051, 0.051, 0.051], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012233587399477745, "centered_acc": -0.08933333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333], "ci": NaN}}, "summary": {"mean_acc_norm": 0.3002600420692572, "mean_acc_norm_ci": NaN, "mean_acc": 0.29872317970911644, "mean_acc_ci": NaN, "mean_centered_acc": 0.010107156495395217, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.3002600420692572, 0.3002600420692572, 0.3002600420692572, 0.3002600420692572, 0.3002600420692572], "mean_acc_runs": [0.29872317970911644, 0.29872317970911644, 0.29872317970911644, 0.29872317970911644, 0.29872317970911644], "mean_centered_acc_runs": [0.010107156495395217, 0.010107156495395217, 0.010107156495395217, 0.010107156495395217, 0.010107156495395217], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
| 2 |
+
{"run_dir": "/workspace/concept_model/runs/i5-exp-fulle-lm-wsd-from-4320", "run": "i5-exp-fulle-lm-wsd-from-4320", "checkpoint_path": "/workspace/concept_model/runs/i5-exp-fulle-lm-wsd-from-4320/checkpoint-4464", "checkpoint": "checkpoint-4464", "global_step": 4464, "tokens_seen": 4680843264, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2544313881696873, "acc_norm_ci": NaN, "acc_norm_runs": [0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873], "acc": 0.2560246962756423, "acc_ci": NaN, "acc_runs": [0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00434650985067984, "centered_acc": 0.005908517559583067, "centered_acc_ci": NaN, "centered_acc_runs": [0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067], "ci": NaN}, "piqa": {"acc_norm": 0.5380848748639826, "acc_norm_ci": NaN, "acc_norm_runs": [0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826], "acc": 0.5457018498367792, "acc_ci": NaN, "acc_runs": [0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011631933367846934, "centered_acc": 0.0761697497279652, "centered_acc_ci": NaN, "centered_acc_runs": [0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652], "ci": NaN}, "arc_easy": {"acc_norm": 0.2777777777777778, "acc_norm_ci": NaN, "acc_norm_runs": [0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778], "acc": 0.26304713804713803, "acc_ci": NaN, "acc_runs": [0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009190779909650122, "centered_acc": 0.037037037037037056, "centered_acc_ci": NaN, "centered_acc_runs": [0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056], "ci": NaN}, "arc_challenge": {"acc_norm": 0.25426621160409557, "acc_norm_ci": NaN, "acc_norm_runs": [0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557], "acc": 0.2090443686006826, "acc_ci": NaN, "acc_runs": [0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012724999945157845, "centered_acc": 0.0056882821387940945, "centered_acc_ci": NaN, "centered_acc_runs": [0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945], "ci": NaN}, "openbookqa": {"acc_norm": 0.294, "acc_norm_ci": NaN, "acc_norm_runs": [0.294, 0.294, 0.294, 0.294, 0.294], "acc": 0.164, "acc_ci": NaN, "acc_runs": [0.164, 0.164, 0.164, 0.164, 0.164], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.02039509548493655, "centered_acc": 0.058666666666666645, "centered_acc_ci": NaN, "centered_acc_runs": [0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5201262825572218, "acc_ci": NaN, "acc_runs": [0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.01404109666434435, "centered_acc": 0.04025256511444364, "centered_acc_ci": NaN, "centered_acc_runs": [0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.6, "acc_ci": NaN, "acc_runs": [0.6, 0.6, 0.6, 0.6, 0.6], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0492365963917331, "centered_acc": 0.19999999999999996, "centered_acc_ci": NaN, "centered_acc_runs": [0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 11127748.30585826, "perplexity_ci": NaN, "perplexity_runs": [11127748.30585826, 11127748.30585826, 11127748.30585826, 11127748.30585826, 11127748.30585826], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.183, "acc_norm_ci": NaN, "acc_norm_runs": [0.183, 0.183, 0.183, 0.183, 0.183], "acc": 0.051000000000000004, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.051, 0.051, 0.051, 0.051, 0.051], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012233587399477745, "centered_acc": -0.08933333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.30026004206925716, "mean_acc_norm_ci": 0.042481836721542376, "mean_acc": 0.309170733758894, "mean_acc_ci": 0.05035993353726309, "mean_centered_acc": 0.014511603300729349, "mean_centered_acc_ci": 0.031496150671264184, "mean_acc_norm_runs": [0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.294, 0.294, 0.294, 0.294, 0.294, 0.183, 0.183, 0.183, 0.183, 0.183], "mean_acc_runs": [0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.164, 0.164, 0.164, 0.164, 0.164, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.6, 0.6, 0.6, 0.6, 0.6, 0.0, 0.0, 0.0, 0.0, 0.0, 0.051, 0.051, 0.051, 0.051, 0.051, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "mean_centered_acc_runs": [0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.042481836721542376}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
| 3 |
+
{"run_dir": "/workspace/concept_model/runs/i5-exp-fulle-lm-wsd-from-4320", "run": "i5-exp-fulle-lm-wsd-from-4320", "checkpoint_path": "/workspace/concept_model/runs/i5-exp-fulle-lm-wsd-from-4320/checkpoint-4464", "checkpoint": "checkpoint-4464", "global_step": 4464, "tokens_seen": 4680843264, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "blimp", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2544313881696873, "acc_norm_ci": NaN, "acc_norm_runs": [0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873], "acc": 0.2560246962756423, "acc_ci": NaN, "acc_runs": [0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00434650985067984, "centered_acc": 0.005908517559583067, "centered_acc_ci": NaN, "centered_acc_runs": [0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067], "ci": NaN}, "piqa": {"acc_norm": 0.5380848748639826, "acc_norm_ci": NaN, "acc_norm_runs": [0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826], "acc": 0.5457018498367792, "acc_ci": NaN, "acc_runs": [0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011631933367846934, "centered_acc": 0.0761697497279652, "centered_acc_ci": NaN, "centered_acc_runs": [0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652], "ci": NaN}, "arc_easy": {"acc_norm": 0.2777777777777778, "acc_norm_ci": NaN, "acc_norm_runs": [0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778], "acc": 0.26304713804713803, "acc_ci": NaN, "acc_runs": [0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009190779909650122, "centered_acc": 0.037037037037037056, "centered_acc_ci": NaN, "centered_acc_runs": [0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056], "ci": NaN}, "arc_challenge": {"acc_norm": 0.25426621160409557, "acc_norm_ci": NaN, "acc_norm_runs": [0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557], "acc": 0.2090443686006826, "acc_ci": NaN, "acc_runs": [0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012724999945157845, "centered_acc": 0.0056882821387940945, "centered_acc_ci": NaN, "centered_acc_runs": [0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945], "ci": NaN}, "openbookqa": {"acc_norm": 0.294, "acc_norm_ci": NaN, "acc_norm_runs": [0.294, 0.294, 0.294, 0.294, 0.294], "acc": 0.164, "acc_ci": NaN, "acc_runs": [0.164, 0.164, 0.164, 0.164, 0.164], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.02039509548493655, "centered_acc": 0.058666666666666645, "centered_acc_ci": NaN, "centered_acc_runs": [0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5201262825572218, "acc_ci": NaN, "acc_runs": [0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.01404109666434435, "centered_acc": 0.04025256511444364, "centered_acc_ci": NaN, "centered_acc_runs": [0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.6, "acc_ci": NaN, "acc_runs": [0.6, 0.6, 0.6, 0.6, 0.6], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0492365963917331, "centered_acc": 0.19999999999999996, "centered_acc_ci": NaN, "centered_acc_runs": [0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 11127748.30585826, "perplexity_ci": NaN, "perplexity_runs": [11127748.30585826, 11127748.30585826, 11127748.30585826, 11127748.30585826, 11127748.30585826], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.183, "acc_norm_ci": NaN, "acc_norm_runs": [0.183, 0.183, 0.183, 0.183, 0.183], "acc": 0.051000000000000004, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.051, 0.051, 0.051, 0.051, 0.051], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012233587399477745, "centered_acc": -0.08933333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "blimp": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5779850746268657, "acc_ci": NaN, "acc_runs": [0.5779850746268657, 0.5779850746268657, 0.5779850746268657, 0.5779850746268657, 0.5779850746268657], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0017650183330752338, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.30026004206925716, "mean_acc_norm_ci": 0.042481836721542376, "mean_acc": 0.3298487599795072, "mean_acc_ci": 0.04970978100318946, "mean_centered_acc": 0.014511603300729349, "mean_centered_acc_ci": 0.031496150671264184, "mean_acc_norm_runs": [0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.2544313881696873, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.5380848748639826, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.2777777777777778, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.294, 0.294, 0.294, 0.294, 0.294, 0.183, 0.183, 0.183, 0.183, 0.183], "mean_acc_runs": [0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.2560246962756423, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.5457018498367792, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.26304713804713803, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.2090443686006826, 0.164, 0.164, 0.164, 0.164, 0.164, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.5201262825572218, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.6, 0.6, 0.6, 0.6, 0.6, 0.0, 0.0, 0.0, 0.0, 0.0, 0.051, 0.051, 0.051, 0.051, 0.051, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.5779850746268657, 0.5779850746268657, 0.5779850746268657, 0.5779850746268657, 0.5779850746268657], "mean_centered_acc_runs": [0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.005908517559583067, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.0761697497279652, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.037037037037037056, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.058666666666666645, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364, 0.04025256511444364, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996, 0.19999999999999996, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.042481836721542376}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
results/l2r50-fulle-lm/zero_shot_eval/checkpoint-2304.json
CHANGED
|
@@ -6,19 +6,21 @@
|
|
| 6 |
"global_step": 2304,
|
| 7 |
"tokens_seen": 2415919104,
|
| 8 |
"num_fewshot": 0,
|
| 9 |
-
"runs_per_task":
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
| 12 |
-
"hellaswag",
|
| 13 |
-
"piqa",
|
| 14 |
-
"arc_easy",
|
| 15 |
"arc_challenge",
|
| 16 |
-
"
|
| 17 |
-
"winogrande",
|
| 18 |
"boolq",
|
|
|
|
| 19 |
"copa",
|
|
|
|
| 20 |
"lambada_openai",
|
| 21 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
],
|
| 23 |
"tasks": {
|
| 24 |
"hellaswag": {
|
|
@@ -232,30 +234,124 @@
|
|
| 232 |
-0.084
|
| 233 |
],
|
| 234 |
"ci": 0.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 235 |
}
|
| 236 |
},
|
| 237 |
"summary": {
|
| 238 |
"mean_acc_norm": 0.29801237851945755,
|
| 239 |
-
"mean_acc_norm_ci": 0.
|
| 240 |
-
"mean_acc": 0.
|
| 241 |
-
"mean_acc_ci": 0.
|
| 242 |
-
"mean_centered_acc":
|
| 243 |
-
"mean_centered_acc_ci": 0.
|
| 244 |
"mean_acc_norm_runs": [
|
| 245 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 246 |
],
|
| 247 |
"mean_acc_runs": [
|
| 248 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 249 |
],
|
| 250 |
"mean_centered_acc_runs": [
|
| 251 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 252 |
],
|
| 253 |
-
"ci": 0.
|
| 254 |
},
|
| 255 |
"batch_size_per_device": 8,
|
| 256 |
"device": "cuda:0",
|
| 257 |
"dtype": "float16",
|
| 258 |
-
"
|
| 259 |
-
"
|
| 260 |
-
"
|
| 261 |
-
|
|
|
|
|
|
|
|
|
| 6 |
"global_step": 2304,
|
| 7 |
"tokens_seen": 2415919104,
|
| 8 |
"num_fewshot": 0,
|
| 9 |
+
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
|
|
|
|
|
|
|
|
|
| 12 |
"arc_challenge",
|
| 13 |
+
"arc_easy",
|
|
|
|
| 14 |
"boolq",
|
| 15 |
+
"commonsense_qa",
|
| 16 |
"copa",
|
| 17 |
+
"hellaswag",
|
| 18 |
"lambada_openai",
|
| 19 |
+
"openbookqa",
|
| 20 |
+
"piqa",
|
| 21 |
+
"rte",
|
| 22 |
+
"sciq",
|
| 23 |
+
"winogrande"
|
| 24 |
],
|
| 25 |
"tasks": {
|
| 26 |
"hellaswag": {
|
|
|
|
| 234 |
-0.084
|
| 235 |
],
|
| 236 |
"ci": 0.0
|
| 237 |
+
},
|
| 238 |
+
"rte": {
|
| 239 |
+
"acc_norm": null,
|
| 240 |
+
"acc_norm_ci": 0.0,
|
| 241 |
+
"acc_norm_runs": [],
|
| 242 |
+
"acc": 0.5270758122743683,
|
| 243 |
+
"acc_ci": NaN,
|
| 244 |
+
"acc_runs": [
|
| 245 |
+
0.5270758122743683,
|
| 246 |
+
0.5270758122743683,
|
| 247 |
+
0.5270758122743683,
|
| 248 |
+
0.5270758122743683,
|
| 249 |
+
0.5270758122743683
|
| 250 |
+
],
|
| 251 |
+
"perplexity": null,
|
| 252 |
+
"perplexity_ci": 0.0,
|
| 253 |
+
"perplexity_runs": [],
|
| 254 |
+
"stderr": 0.030052303463143738,
|
| 255 |
+
"centered_acc": 0.054151624548736566,
|
| 256 |
+
"centered_acc_ci": NaN,
|
| 257 |
+
"centered_acc_runs": [
|
| 258 |
+
0.054151624548736566,
|
| 259 |
+
0.054151624548736566,
|
| 260 |
+
0.054151624548736566,
|
| 261 |
+
0.054151624548736566,
|
| 262 |
+
0.054151624548736566
|
| 263 |
+
],
|
| 264 |
+
"ci": NaN
|
| 265 |
+
},
|
| 266 |
+
"commonsense_qa": {
|
| 267 |
+
"acc_norm": null,
|
| 268 |
+
"acc_norm_ci": 0.0,
|
| 269 |
+
"acc_norm_runs": [],
|
| 270 |
+
"acc": 0.19574119574119575,
|
| 271 |
+
"acc_ci": NaN,
|
| 272 |
+
"acc_runs": [
|
| 273 |
+
0.19574119574119575,
|
| 274 |
+
0.19574119574119575,
|
| 275 |
+
0.19574119574119575,
|
| 276 |
+
0.19574119574119575,
|
| 277 |
+
0.19574119574119575
|
| 278 |
+
],
|
| 279 |
+
"perplexity": null,
|
| 280 |
+
"perplexity_ci": 0.0,
|
| 281 |
+
"perplexity_runs": [],
|
| 282 |
+
"stderr": 0.011359497363584397,
|
| 283 |
+
"centered_acc": null,
|
| 284 |
+
"centered_acc_ci": 0.0,
|
| 285 |
+
"centered_acc_runs": [],
|
| 286 |
+
"ci": NaN
|
| 287 |
}
|
| 288 |
},
|
| 289 |
"summary": {
|
| 290 |
"mean_acc_norm": 0.29801237851945755,
|
| 291 |
+
"mean_acc_norm_ci": 0.12931242637195622,
|
| 292 |
+
"mean_acc": 0.3685919670287551,
|
| 293 |
+
"mean_acc_ci": 0.07639876669218759,
|
| 294 |
+
"mean_centered_acc": 0.018916383006119298,
|
| 295 |
+
"mean_centered_acc_ci": 0.05103601272583667,
|
| 296 |
"mean_acc_norm_runs": [
|
| 297 |
+
0.2561242780322645,
|
| 298 |
+
0.5397170837867247,
|
| 299 |
+
0.2689393939393939,
|
| 300 |
+
0.24829351535836178,
|
| 301 |
+
0.288,
|
| 302 |
+
0.187
|
| 303 |
],
|
| 304 |
"mean_acc_runs": [
|
| 305 |
+
0.2560246962756423,
|
| 306 |
+
0.5642002176278563,
|
| 307 |
+
0.25757575757575757,
|
| 308 |
+
0.22184300341296928,
|
| 309 |
+
0.164,
|
| 310 |
+
0.5201262825572218,
|
| 311 |
+
0.3782874617737003,
|
| 312 |
+
0.56,
|
| 313 |
+
0.0,
|
| 314 |
+
0.05,
|
| 315 |
+
0.5270758122743683,
|
| 316 |
+
0.5270758122743683,
|
| 317 |
+
0.5270758122743683,
|
| 318 |
+
0.5270758122743683,
|
| 319 |
+
0.5270758122743683,
|
| 320 |
+
0.19574119574119575,
|
| 321 |
+
0.19574119574119575,
|
| 322 |
+
0.19574119574119575,
|
| 323 |
+
0.19574119574119575,
|
| 324 |
+
0.19574119574119575,
|
| 325 |
+
0.5257313432835821,
|
| 326 |
+
0.5257313432835821,
|
| 327 |
+
0.5257313432835821,
|
| 328 |
+
0.5257313432835821,
|
| 329 |
+
0.5257313432835821
|
| 330 |
],
|
| 331 |
"mean_centered_acc_runs": [
|
| 332 |
+
0.008165704043019343,
|
| 333 |
+
0.07943416757344934,
|
| 334 |
+
0.02525252525252523,
|
| 335 |
+
-0.0022753128555176305,
|
| 336 |
+
0.05066666666666664,
|
| 337 |
+
0.04025256511444364,
|
| 338 |
+
-0.24342507645259936,
|
| 339 |
+
0.1200000000000001,
|
| 340 |
+
-0.084,
|
| 341 |
+
0.054151624548736566,
|
| 342 |
+
0.054151624548736566,
|
| 343 |
+
0.054151624548736566,
|
| 344 |
+
0.054151624548736566,
|
| 345 |
+
0.054151624548736566
|
| 346 |
],
|
| 347 |
+
"ci": 0.12931242637195622
|
| 348 |
},
|
| 349 |
"batch_size_per_device": 8,
|
| 350 |
"device": "cuda:0",
|
| 351 |
"dtype": "float16",
|
| 352 |
+
"add_l2r_token": true,
|
| 353 |
+
"add_next_1_pred_token": false,
|
| 354 |
+
"tokenizer_has_l2r_pred": true,
|
| 355 |
+
"tokenizer_has_next_1_pred": false,
|
| 356 |
+
"max_next_i_train": 1
|
| 357 |
+
}
|
results/l2r50-fulle-lm/zero_shot_eval/summary.jsonl
CHANGED
|
@@ -31,3 +31,5 @@
|
|
| 31 |
{"run_dir": "/workspace/concept_model/runs/l2r50-fulle-lm", "run": "l2r50-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/l2r50-fulle-lm/checkpoint-8640", "checkpoint": "checkpoint-8640", "global_step": 8640, "tokens_seen": 9059696640, "num_fewshot": 0, "runs_per_task": 1, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2574188408683529, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2574188408683529], "acc": 0.25413264289982074, "acc_ci": 0.0, "acc_runs": [0.25413264289982074], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0043631851720466585, "centered_acc": 0.009891787824470569, "centered_acc_ci": 0.0, "centered_acc_runs": [0.009891787824470569], "ci": 0.0}, "piqa": {"acc_norm": 0.5217627856365615, "acc_norm_ci": 0.0, "acc_norm_runs": [0.5217627856365615], "acc": 0.528835690968444, "acc_ci": 0.0, "acc_runs": [0.528835690968444], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011654768618559977, "centered_acc": 0.043525571273123065, "centered_acc_ci": 0.0, "centered_acc_runs": [0.043525571273123065], "ci": 0.0}, "arc_easy": {"acc_norm": 0.2702020202020202, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2702020202020202], "acc": 0.25715488215488214, "acc_ci": 0.0, "acc_runs": [0.25715488215488214], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009112002229119622, "centered_acc": 0.026936026936026963, "centered_acc_ci": 0.0, "centered_acc_runs": [0.026936026936026963], "ci": 0.0}, "arc_challenge": {"acc_norm": 0.25, "acc_norm_ci": 0.0, "acc_norm_runs": [0.25], "acc": 0.21160409556313994, "acc_ci": 0.0, "acc_runs": [0.21160409556313994], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012653835621466646, "centered_acc": 0.0, "centered_acc_ci": 0.0, "centered_acc_runs": [0.0], "ci": 0.0}, "openbookqa": {"acc_norm": 0.278, "acc_norm_ci": 0.0, "acc_norm_runs": [0.278], "acc": 0.166, "acc_ci": 0.0, "acc_runs": [0.166], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020055833888070865, "centered_acc": 0.037333333333333364, "centered_acc_ci": 0.0, "centered_acc_runs": [0.037333333333333364], "ci": 0.0}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5209155485398579, "acc_ci": 0.0, "acc_runs": [0.5209155485398579], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014040185494212997, "centered_acc": 0.04183109707971577, "centered_acc_ci": 0.0, "centered_acc_runs": [0.04183109707971577], "ci": 0.0}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": 0.0, "acc_runs": [0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.24342507645259936], "ci": 0.0}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.58, "acc_ci": 0.0, "acc_runs": [0.58], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.04960449637488582, "centered_acc": 0.15999999999999992, "centered_acc_ci": 0.0, "centered_acc_runs": [0.15999999999999992], "ci": 0.0}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": 0.0, "acc_runs": [0.0], "perplexity": 130194073.54691643, "perplexity_ci": 0.0, "perplexity_runs": [130194073.54691643], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": 0.0}, "sciq": {"acc_norm": 0.18, "acc_norm_ci": 0.0, "acc_norm_runs": [0.18], "acc": 0.047, "acc_ci": 0.0, "acc_runs": [0.047], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012155153135512024, "centered_acc": -0.09333333333333334, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.09333333333333334], "ci": 0.0}}, "summary": {"mean_acc_norm": 0.2928972744511558, "mean_acc_norm_ci": 0.0, "mean_acc": 0.29439303218998447, "mean_acc_ci": 0.0, "mean_centered_acc": -0.00191562148214034, "mean_centered_acc_ci": 0.0, "mean_acc_norm_runs": [0.2928972744511558], "mean_acc_runs": [0.29439303218998447], "mean_centered_acc_runs": [-0.00191562148214034], "ci": 0.0}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "next_i_enabled": false, "max_next_i_train": 1, "l2r_prefix_enabled": true}
|
| 32 |
{"run_dir": "/workspace/concept_model/runs/l2r50-fulle-lm", "run": "l2r50-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/l2r50-fulle-lm/checkpoint-8928", "checkpoint": "checkpoint-8928", "global_step": 8928, "tokens_seen": 9361686528, "num_fewshot": 0, "runs_per_task": 1, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.25801633140808605, "acc_norm_ci": 0.0, "acc_norm_runs": [0.25801633140808605], "acc": 0.2555267874925314, "acc_ci": 0.0, "acc_runs": [0.2555267874925314], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004366488167386302, "centered_acc": 0.01068844187744807, "centered_acc_ci": 0.0, "centered_acc_runs": [0.01068844187744807], "ci": 0.0}, "piqa": {"acc_norm": 0.528835690968444, "acc_norm_ci": 0.0, "acc_norm_runs": [0.528835690968444], "acc": 0.5484221980413493, "acc_ci": 0.0, "acc_runs": [0.5484221980413493], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011646407809944788, "centered_acc": 0.05767138193688792, "centered_acc_ci": 0.0, "centered_acc_runs": [0.05767138193688792], "ci": 0.0}, "arc_easy": {"acc_norm": 0.2748316498316498, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2748316498316498], "acc": 0.25757575757575757, "acc_ci": 0.0, "acc_runs": [0.25757575757575757], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009160538115254864, "centered_acc": 0.03310886644219976, "centered_acc_ci": 0.0, "centered_acc_runs": [0.03310886644219976], "ci": 0.0}, "arc_challenge": {"acc_norm": 0.24914675767918087, "acc_norm_ci": 0.0, "acc_norm_runs": [0.24914675767918087], "acc": 0.20477815699658702, "acc_ci": 0.0, "acc_runs": [0.20477815699658702], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012639407111926505, "centered_acc": -0.0011376564277588337, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.0011376564277588337], "ci": 0.0}, "openbookqa": {"acc_norm": 0.278, "acc_norm_ci": 0.0, "acc_norm_runs": [0.278], "acc": 0.168, "acc_ci": 0.0, "acc_runs": [0.168], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020055833888070865, "centered_acc": 0.037333333333333364, "centered_acc_ci": 0.0, "centered_acc_runs": [0.037333333333333364], "ci": 0.0}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5217048145224941, "acc_ci": 0.0, "acc_runs": [0.5217048145224941], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014039239216484667, "centered_acc": 0.04340962904498813, "centered_acc_ci": 0.0, "centered_acc_runs": [0.04340962904498813], "ci": 0.0}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": 0.0, "acc_runs": [0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.24342507645259936], "ci": 0.0}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.57, "acc_ci": 0.0, "acc_runs": [0.57], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.049756985195624305, "centered_acc": 0.1399999999999999, "centered_acc_ci": 0.0, "centered_acc_runs": [0.1399999999999999], "ci": 0.0}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": 0.0, "acc_runs": [0.0], "perplexity": 119406792.69935393, "perplexity_ci": 0.0, "perplexity_runs": [119406792.69935393], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": 0.0}, "sciq": {"acc_norm": 0.182, "acc_norm_ci": 0.0, "acc_norm_runs": [0.182], "acc": 0.05, "acc_ci": 0.0, "acc_runs": [0.05], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0122075806376622, "centered_acc": -0.09066666666666667, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.09066666666666667], "ci": 0.0}}, "summary": {"mean_acc_norm": 0.2951384049812268, "mean_acc_norm_ci": 0.0, "mean_acc": 0.29542951764024195, "mean_acc_ci": 0.0, "mean_centered_acc": -0.0014464163235741933, "mean_centered_acc_ci": 0.0, "mean_acc_norm_runs": [0.2951384049812268], "mean_acc_runs": [0.29542951764024195], "mean_centered_acc_runs": [-0.0014464163235741933], "ci": 0.0}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "next_i_enabled": false, "max_next_i_train": 1, "l2r_prefix_enabled": true}
|
| 33 |
{"run_dir": "/workspace/concept_model/runs/l2r50-fulle-lm", "run": "l2r50-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/l2r50-fulle-lm/checkpoint-9216", "checkpoint": "checkpoint-9216", "global_step": 9216, "tokens_seen": 9663676416, "num_fewshot": 0, "runs_per_task": 1, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2562238597888867, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2562238597888867], "acc": 0.2570205138418642, "acc_ci": 0.0, "acc_runs": [0.2570205138418642], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004356547185846709, "centered_acc": 0.00829847971851557, "centered_acc_ci": 0.0, "centered_acc_runs": [0.00829847971851557], "ci": 0.0}, "piqa": {"acc_norm": 0.5195865070729053, "acc_norm_ci": 0.0, "acc_norm_runs": [0.5195865070729053], "acc": 0.5310119695321001, "acc_ci": 0.0, "acc_runs": [0.5310119695321001], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.01165686997928841, "centered_acc": 0.03917301414581065, "centered_acc_ci": 0.0, "centered_acc_runs": [0.03917301414581065], "ci": 0.0}, "arc_easy": {"acc_norm": 0.265993265993266, "acc_norm_ci": 0.0, "acc_norm_runs": [0.265993265993266], "acc": 0.2558922558922559, "acc_ci": 0.0, "acc_runs": [0.2558922558922559], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009066789565615746, "centered_acc": 0.021324354657688005, "centered_acc_ci": 0.0, "centered_acc_runs": [0.021324354657688005], "ci": 0.0}, "arc_challenge": {"acc_norm": 0.2508532423208191, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2508532423208191], "acc": 0.21331058020477817, "acc_ci": 0.0, "acc_runs": [0.21331058020477817], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012668198621315347, "centered_acc": 0.0011376564277588337, "centered_acc_ci": 0.0, "centered_acc_runs": [0.0011376564277588337], "ci": 0.0}, "openbookqa": {"acc_norm": 0.276, "acc_norm_ci": 0.0, "acc_norm_runs": [0.276], "acc": 0.172, "acc_ci": 0.0, "acc_runs": [0.172], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.02001121929807355, "centered_acc": 0.0346666666666667, "centered_acc_ci": 0.0, "centered_acc_runs": [0.0346666666666667], "ci": 0.0}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.526440410418311, "acc_ci": 0.0, "acc_runs": [0.526440410418311], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014032823874407293, "centered_acc": 0.052880820836622044, "centered_acc_ci": 0.0, "centered_acc_runs": [0.052880820836622044], "ci": 0.0}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": 0.0, "acc_runs": [0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.24342507645259936], "ci": 0.0}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.55, "acc_ci": 0.0, "acc_runs": [0.55], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05, "centered_acc": 0.10000000000000009, "centered_acc_ci": 0.0, "centered_acc_runs": [0.10000000000000009], "ci": 0.0}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": 0.0, "acc_runs": [0.0], "perplexity": 122033921.27246587, "perplexity_ci": 0.0, "perplexity_runs": [122033921.27246587], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": 0.0}, "sciq": {"acc_norm": 0.182, "acc_norm_ci": 0.0, "acc_norm_runs": [0.182], "acc": 0.052, "acc_ci": 0.0, "acc_runs": [0.052], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0122075806376622, "centered_acc": -0.09066666666666667, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.09066666666666667], "ci": 0.0}}, "summary": {"mean_acc_norm": 0.29177614586264616, "mean_acc_norm_ci": 0.0, "mean_acc": 0.29359631916630097, "mean_acc_ci": 0.0, "mean_centered_acc": -0.008512305629578238, "mean_centered_acc_ci": 0.0, "mean_acc_norm_runs": [0.29177614586264616], "mean_acc_runs": [0.29359631916630097], "mean_centered_acc_runs": [-0.008512305629578238], "ci": 0.0}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "next_i_enabled": false, "max_next_i_train": 1, "l2r_prefix_enabled": true}
|
|
|
|
|
|
|
|
|
| 31 |
{"run_dir": "/workspace/concept_model/runs/l2r50-fulle-lm", "run": "l2r50-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/l2r50-fulle-lm/checkpoint-8640", "checkpoint": "checkpoint-8640", "global_step": 8640, "tokens_seen": 9059696640, "num_fewshot": 0, "runs_per_task": 1, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2574188408683529, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2574188408683529], "acc": 0.25413264289982074, "acc_ci": 0.0, "acc_runs": [0.25413264289982074], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0043631851720466585, "centered_acc": 0.009891787824470569, "centered_acc_ci": 0.0, "centered_acc_runs": [0.009891787824470569], "ci": 0.0}, "piqa": {"acc_norm": 0.5217627856365615, "acc_norm_ci": 0.0, "acc_norm_runs": [0.5217627856365615], "acc": 0.528835690968444, "acc_ci": 0.0, "acc_runs": [0.528835690968444], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011654768618559977, "centered_acc": 0.043525571273123065, "centered_acc_ci": 0.0, "centered_acc_runs": [0.043525571273123065], "ci": 0.0}, "arc_easy": {"acc_norm": 0.2702020202020202, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2702020202020202], "acc": 0.25715488215488214, "acc_ci": 0.0, "acc_runs": [0.25715488215488214], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009112002229119622, "centered_acc": 0.026936026936026963, "centered_acc_ci": 0.0, "centered_acc_runs": [0.026936026936026963], "ci": 0.0}, "arc_challenge": {"acc_norm": 0.25, "acc_norm_ci": 0.0, "acc_norm_runs": [0.25], "acc": 0.21160409556313994, "acc_ci": 0.0, "acc_runs": [0.21160409556313994], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012653835621466646, "centered_acc": 0.0, "centered_acc_ci": 0.0, "centered_acc_runs": [0.0], "ci": 0.0}, "openbookqa": {"acc_norm": 0.278, "acc_norm_ci": 0.0, "acc_norm_runs": [0.278], "acc": 0.166, "acc_ci": 0.0, "acc_runs": [0.166], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020055833888070865, "centered_acc": 0.037333333333333364, "centered_acc_ci": 0.0, "centered_acc_runs": [0.037333333333333364], "ci": 0.0}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5209155485398579, "acc_ci": 0.0, "acc_runs": [0.5209155485398579], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014040185494212997, "centered_acc": 0.04183109707971577, "centered_acc_ci": 0.0, "centered_acc_runs": [0.04183109707971577], "ci": 0.0}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": 0.0, "acc_runs": [0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.24342507645259936], "ci": 0.0}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.58, "acc_ci": 0.0, "acc_runs": [0.58], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.04960449637488582, "centered_acc": 0.15999999999999992, "centered_acc_ci": 0.0, "centered_acc_runs": [0.15999999999999992], "ci": 0.0}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": 0.0, "acc_runs": [0.0], "perplexity": 130194073.54691643, "perplexity_ci": 0.0, "perplexity_runs": [130194073.54691643], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": 0.0}, "sciq": {"acc_norm": 0.18, "acc_norm_ci": 0.0, "acc_norm_runs": [0.18], "acc": 0.047, "acc_ci": 0.0, "acc_runs": [0.047], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012155153135512024, "centered_acc": -0.09333333333333334, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.09333333333333334], "ci": 0.0}}, "summary": {"mean_acc_norm": 0.2928972744511558, "mean_acc_norm_ci": 0.0, "mean_acc": 0.29439303218998447, "mean_acc_ci": 0.0, "mean_centered_acc": -0.00191562148214034, "mean_centered_acc_ci": 0.0, "mean_acc_norm_runs": [0.2928972744511558], "mean_acc_runs": [0.29439303218998447], "mean_centered_acc_runs": [-0.00191562148214034], "ci": 0.0}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "next_i_enabled": false, "max_next_i_train": 1, "l2r_prefix_enabled": true}
|
| 32 |
{"run_dir": "/workspace/concept_model/runs/l2r50-fulle-lm", "run": "l2r50-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/l2r50-fulle-lm/checkpoint-8928", "checkpoint": "checkpoint-8928", "global_step": 8928, "tokens_seen": 9361686528, "num_fewshot": 0, "runs_per_task": 1, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.25801633140808605, "acc_norm_ci": 0.0, "acc_norm_runs": [0.25801633140808605], "acc": 0.2555267874925314, "acc_ci": 0.0, "acc_runs": [0.2555267874925314], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004366488167386302, "centered_acc": 0.01068844187744807, "centered_acc_ci": 0.0, "centered_acc_runs": [0.01068844187744807], "ci": 0.0}, "piqa": {"acc_norm": 0.528835690968444, "acc_norm_ci": 0.0, "acc_norm_runs": [0.528835690968444], "acc": 0.5484221980413493, "acc_ci": 0.0, "acc_runs": [0.5484221980413493], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011646407809944788, "centered_acc": 0.05767138193688792, "centered_acc_ci": 0.0, "centered_acc_runs": [0.05767138193688792], "ci": 0.0}, "arc_easy": {"acc_norm": 0.2748316498316498, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2748316498316498], "acc": 0.25757575757575757, "acc_ci": 0.0, "acc_runs": [0.25757575757575757], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009160538115254864, "centered_acc": 0.03310886644219976, "centered_acc_ci": 0.0, "centered_acc_runs": [0.03310886644219976], "ci": 0.0}, "arc_challenge": {"acc_norm": 0.24914675767918087, "acc_norm_ci": 0.0, "acc_norm_runs": [0.24914675767918087], "acc": 0.20477815699658702, "acc_ci": 0.0, "acc_runs": [0.20477815699658702], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012639407111926505, "centered_acc": -0.0011376564277588337, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.0011376564277588337], "ci": 0.0}, "openbookqa": {"acc_norm": 0.278, "acc_norm_ci": 0.0, "acc_norm_runs": [0.278], "acc": 0.168, "acc_ci": 0.0, "acc_runs": [0.168], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020055833888070865, "centered_acc": 0.037333333333333364, "centered_acc_ci": 0.0, "centered_acc_runs": [0.037333333333333364], "ci": 0.0}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5217048145224941, "acc_ci": 0.0, "acc_runs": [0.5217048145224941], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014039239216484667, "centered_acc": 0.04340962904498813, "centered_acc_ci": 0.0, "centered_acc_runs": [0.04340962904498813], "ci": 0.0}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": 0.0, "acc_runs": [0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.24342507645259936], "ci": 0.0}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.57, "acc_ci": 0.0, "acc_runs": [0.57], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.049756985195624305, "centered_acc": 0.1399999999999999, "centered_acc_ci": 0.0, "centered_acc_runs": [0.1399999999999999], "ci": 0.0}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": 0.0, "acc_runs": [0.0], "perplexity": 119406792.69935393, "perplexity_ci": 0.0, "perplexity_runs": [119406792.69935393], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": 0.0}, "sciq": {"acc_norm": 0.182, "acc_norm_ci": 0.0, "acc_norm_runs": [0.182], "acc": 0.05, "acc_ci": 0.0, "acc_runs": [0.05], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0122075806376622, "centered_acc": -0.09066666666666667, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.09066666666666667], "ci": 0.0}}, "summary": {"mean_acc_norm": 0.2951384049812268, "mean_acc_norm_ci": 0.0, "mean_acc": 0.29542951764024195, "mean_acc_ci": 0.0, "mean_centered_acc": -0.0014464163235741933, "mean_centered_acc_ci": 0.0, "mean_acc_norm_runs": [0.2951384049812268], "mean_acc_runs": [0.29542951764024195], "mean_centered_acc_runs": [-0.0014464163235741933], "ci": 0.0}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "next_i_enabled": false, "max_next_i_train": 1, "l2r_prefix_enabled": true}
|
| 33 |
{"run_dir": "/workspace/concept_model/runs/l2r50-fulle-lm", "run": "l2r50-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/l2r50-fulle-lm/checkpoint-9216", "checkpoint": "checkpoint-9216", "global_step": 9216, "tokens_seen": 9663676416, "num_fewshot": 0, "runs_per_task": 1, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2562238597888867, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2562238597888867], "acc": 0.2570205138418642, "acc_ci": 0.0, "acc_runs": [0.2570205138418642], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004356547185846709, "centered_acc": 0.00829847971851557, "centered_acc_ci": 0.0, "centered_acc_runs": [0.00829847971851557], "ci": 0.0}, "piqa": {"acc_norm": 0.5195865070729053, "acc_norm_ci": 0.0, "acc_norm_runs": [0.5195865070729053], "acc": 0.5310119695321001, "acc_ci": 0.0, "acc_runs": [0.5310119695321001], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.01165686997928841, "centered_acc": 0.03917301414581065, "centered_acc_ci": 0.0, "centered_acc_runs": [0.03917301414581065], "ci": 0.0}, "arc_easy": {"acc_norm": 0.265993265993266, "acc_norm_ci": 0.0, "acc_norm_runs": [0.265993265993266], "acc": 0.2558922558922559, "acc_ci": 0.0, "acc_runs": [0.2558922558922559], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009066789565615746, "centered_acc": 0.021324354657688005, "centered_acc_ci": 0.0, "centered_acc_runs": [0.021324354657688005], "ci": 0.0}, "arc_challenge": {"acc_norm": 0.2508532423208191, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2508532423208191], "acc": 0.21331058020477817, "acc_ci": 0.0, "acc_runs": [0.21331058020477817], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012668198621315347, "centered_acc": 0.0011376564277588337, "centered_acc_ci": 0.0, "centered_acc_runs": [0.0011376564277588337], "ci": 0.0}, "openbookqa": {"acc_norm": 0.276, "acc_norm_ci": 0.0, "acc_norm_runs": [0.276], "acc": 0.172, "acc_ci": 0.0, "acc_runs": [0.172], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.02001121929807355, "centered_acc": 0.0346666666666667, "centered_acc_ci": 0.0, "centered_acc_runs": [0.0346666666666667], "ci": 0.0}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.526440410418311, "acc_ci": 0.0, "acc_runs": [0.526440410418311], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014032823874407293, "centered_acc": 0.052880820836622044, "centered_acc_ci": 0.0, "centered_acc_runs": [0.052880820836622044], "ci": 0.0}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": 0.0, "acc_runs": [0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.24342507645259936], "ci": 0.0}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.55, "acc_ci": 0.0, "acc_runs": [0.55], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05, "centered_acc": 0.10000000000000009, "centered_acc_ci": 0.0, "centered_acc_runs": [0.10000000000000009], "ci": 0.0}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": 0.0, "acc_runs": [0.0], "perplexity": 122033921.27246587, "perplexity_ci": 0.0, "perplexity_runs": [122033921.27246587], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": 0.0}, "sciq": {"acc_norm": 0.182, "acc_norm_ci": 0.0, "acc_norm_runs": [0.182], "acc": 0.052, "acc_ci": 0.0, "acc_runs": [0.052], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0122075806376622, "centered_acc": -0.09066666666666667, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.09066666666666667], "ci": 0.0}}, "summary": {"mean_acc_norm": 0.29177614586264616, "mean_acc_norm_ci": 0.0, "mean_acc": 0.29359631916630097, "mean_acc_ci": 0.0, "mean_centered_acc": -0.008512305629578238, "mean_centered_acc_ci": 0.0, "mean_acc_norm_runs": [0.29177614586264616], "mean_acc_runs": [0.29359631916630097], "mean_centered_acc_runs": [-0.008512305629578238], "ci": 0.0}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "next_i_enabled": false, "max_next_i_train": 1, "l2r_prefix_enabled": true}
|
| 34 |
+
{"run_dir": "/workspace/concept_model/runs/l2r50-fulle-lm", "run": "l2r50-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/l2r50-fulle-lm/checkpoint-2304", "checkpoint": "checkpoint-2304", "global_step": 2304, "tokens_seen": 2415919104, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2561242780322645, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2561242780322645], "acc": 0.2560246962756423, "acc_ci": 0.0, "acc_runs": [0.2560246962756423], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004355992090030594, "centered_acc": 0.008165704043019343, "centered_acc_ci": 0.0, "centered_acc_runs": [0.008165704043019343], "ci": 0.0}, "piqa": {"acc_norm": 0.5397170837867247, "acc_norm_ci": 0.0, "acc_norm_runs": [0.5397170837867247], "acc": 0.5642002176278563, "acc_ci": 0.0, "acc_runs": [0.5642002176278563], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011628961491718543, "centered_acc": 0.07943416757344934, "centered_acc_ci": 0.0, "centered_acc_runs": [0.07943416757344934], "ci": 0.0}, "arc_easy": {"acc_norm": 0.2689393939393939, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2689393939393939], "acc": 0.25757575757575757, "acc_ci": 0.0, "acc_runs": [0.25757575757575757], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009098548093009251, "centered_acc": 0.02525252525252523, "centered_acc_ci": 0.0, "centered_acc_runs": [0.02525252525252523], "ci": 0.0}, "arc_challenge": {"acc_norm": 0.24829351535836178, "acc_norm_ci": 0.0, "acc_norm_runs": [0.24829351535836178], "acc": 0.22184300341296928, "acc_ci": 0.0, "acc_runs": [0.22184300341296928], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012624912868089679, "centered_acc": -0.0022753128555176305, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.0022753128555176305], "ci": 0.0}, "openbookqa": {"acc_norm": 0.288, "acc_norm_ci": 0.0, "acc_norm_runs": [0.288], "acc": 0.164, "acc_ci": 0.0, "acc_runs": [0.164], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020271503835075255, "centered_acc": 0.05066666666666664, "centered_acc_ci": 0.0, "centered_acc_runs": [0.05066666666666664], "ci": 0.0}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5201262825572218, "acc_ci": 0.0, "acc_runs": [0.5201262825572218], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014041096664344348, "centered_acc": 0.04025256511444364, "centered_acc_ci": 0.0, "centered_acc_runs": [0.04025256511444364], "ci": 0.0}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": 0.0, "acc_runs": [0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.24342507645259936], "ci": 0.0}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.56, "acc_ci": 0.0, "acc_runs": [0.56], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.049888765156985884, "centered_acc": 0.1200000000000001, "centered_acc_ci": 0.0, "centered_acc_runs": [0.1200000000000001], "ci": 0.0}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": 0.0, "acc_runs": [0.0], "perplexity": 16434928.321614476, "perplexity_ci": 0.0, "perplexity_runs": [16434928.321614476], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": 0.0}, "sciq": {"acc_norm": 0.187, "acc_norm_ci": 0.0, "acc_norm_runs": [0.187], "acc": 0.05, "acc_ci": 0.0, "acc_runs": [0.05], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012336254828074166, "centered_acc": -0.084, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.084], "ci": 0.0}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.29801237851945755, "mean_acc_norm_ci": 0.12931242637195622, "mean_acc": 0.3293071229650484, "mean_acc_ci": 0.08774246252259099, "mean_centered_acc": 0.018916383006119298, "mean_centered_acc_ci": 0.05103601272583667, "mean_acc_norm_runs": [0.2561242780322645, 0.5397170837867247, 0.2689393939393939, 0.24829351535836178, 0.288, 0.187], "mean_acc_runs": [0.2560246962756423, 0.5642002176278563, 0.25757575757575757, 0.22184300341296928, 0.164, 0.5201262825572218, 0.3782874617737003, 0.56, 0.0, 0.05, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "mean_centered_acc_runs": [0.008165704043019343, 0.07943416757344934, 0.02525252525252523, -0.0022753128555176305, 0.05066666666666664, 0.04025256511444364, -0.24342507645259936, 0.1200000000000001, -0.084, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.12931242637195622}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
| 35 |
+
{"run_dir": "/workspace/concept_model/runs/l2r50-fulle-lm", "run": "l2r50-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/l2r50-fulle-lm/checkpoint-2304", "checkpoint": "checkpoint-2304", "global_step": 2304, "tokens_seen": 2415919104, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "blimp", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2561242780322645, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2561242780322645], "acc": 0.2560246962756423, "acc_ci": 0.0, "acc_runs": [0.2560246962756423], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004355992090030594, "centered_acc": 0.008165704043019343, "centered_acc_ci": 0.0, "centered_acc_runs": [0.008165704043019343], "ci": 0.0}, "piqa": {"acc_norm": 0.5397170837867247, "acc_norm_ci": 0.0, "acc_norm_runs": [0.5397170837867247], "acc": 0.5642002176278563, "acc_ci": 0.0, "acc_runs": [0.5642002176278563], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011628961491718543, "centered_acc": 0.07943416757344934, "centered_acc_ci": 0.0, "centered_acc_runs": [0.07943416757344934], "ci": 0.0}, "arc_easy": {"acc_norm": 0.2689393939393939, "acc_norm_ci": 0.0, "acc_norm_runs": [0.2689393939393939], "acc": 0.25757575757575757, "acc_ci": 0.0, "acc_runs": [0.25757575757575757], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009098548093009251, "centered_acc": 0.02525252525252523, "centered_acc_ci": 0.0, "centered_acc_runs": [0.02525252525252523], "ci": 0.0}, "arc_challenge": {"acc_norm": 0.24829351535836178, "acc_norm_ci": 0.0, "acc_norm_runs": [0.24829351535836178], "acc": 0.22184300341296928, "acc_ci": 0.0, "acc_runs": [0.22184300341296928], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012624912868089679, "centered_acc": -0.0022753128555176305, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.0022753128555176305], "ci": 0.0}, "openbookqa": {"acc_norm": 0.288, "acc_norm_ci": 0.0, "acc_norm_runs": [0.288], "acc": 0.164, "acc_ci": 0.0, "acc_runs": [0.164], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020271503835075255, "centered_acc": 0.05066666666666664, "centered_acc_ci": 0.0, "centered_acc_runs": [0.05066666666666664], "ci": 0.0}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5201262825572218, "acc_ci": 0.0, "acc_runs": [0.5201262825572218], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014041096664344348, "centered_acc": 0.04025256511444364, "centered_acc_ci": 0.0, "centered_acc_runs": [0.04025256511444364], "ci": 0.0}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": 0.0, "acc_runs": [0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.24342507645259936], "ci": 0.0}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.56, "acc_ci": 0.0, "acc_runs": [0.56], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.049888765156985884, "centered_acc": 0.1200000000000001, "centered_acc_ci": 0.0, "centered_acc_runs": [0.1200000000000001], "ci": 0.0}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": 0.0, "acc_runs": [0.0], "perplexity": 16434928.321614476, "perplexity_ci": 0.0, "perplexity_runs": [16434928.321614476], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": 0.0}, "sciq": {"acc_norm": 0.187, "acc_norm_ci": 0.0, "acc_norm_runs": [0.187], "acc": 0.05, "acc_ci": 0.0, "acc_runs": [0.05], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012336254828074166, "centered_acc": -0.084, "centered_acc_ci": 0.0, "centered_acc_runs": [-0.084], "ci": 0.0}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "blimp": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5257313432835821, "acc_ci": NaN, "acc_runs": [0.5257313432835821, 0.5257313432835821, 0.5257313432835821, 0.5257313432835821, 0.5257313432835821], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0017807500739245648, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.29801237851945755, "mean_acc_norm_ci": 0.12931242637195622, "mean_acc": 0.3685919670287551, "mean_acc_ci": 0.07639876669218759, "mean_centered_acc": 0.018916383006119298, "mean_centered_acc_ci": 0.05103601272583667, "mean_acc_norm_runs": [0.2561242780322645, 0.5397170837867247, 0.2689393939393939, 0.24829351535836178, 0.288, 0.187], "mean_acc_runs": [0.2560246962756423, 0.5642002176278563, 0.25757575757575757, 0.22184300341296928, 0.164, 0.5201262825572218, 0.3782874617737003, 0.56, 0.0, 0.05, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.5257313432835821, 0.5257313432835821, 0.5257313432835821, 0.5257313432835821, 0.5257313432835821], "mean_centered_acc_runs": [0.008165704043019343, 0.07943416757344934, 0.02525252525252523, -0.0022753128555176305, 0.05066666666666664, 0.04025256511444364, -0.24342507645259936, 0.1200000000000001, -0.084, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.12931242637195622}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
results/l2r50-i5-exp-fulle-lm-wsd-from-3168/zero_shot_eval/checkpoint-3312.json
CHANGED
|
@@ -9,16 +9,18 @@
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
| 12 |
-
"hellaswag",
|
| 13 |
-
"piqa",
|
| 14 |
-
"arc_easy",
|
| 15 |
"arc_challenge",
|
| 16 |
-
"
|
| 17 |
-
"winogrande",
|
| 18 |
"boolq",
|
|
|
|
| 19 |
"copa",
|
|
|
|
| 20 |
"lambada_openai",
|
| 21 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
],
|
| 23 |
"tasks": {
|
| 24 |
"hellaswag": {
|
|
@@ -336,37 +338,217 @@
|
|
| 336 |
-0.08933333333333333
|
| 337 |
],
|
| 338 |
"ci": NaN
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 339 |
}
|
| 340 |
},
|
| 341 |
"summary": {
|
| 342 |
-
"mean_acc_norm": 0.
|
| 343 |
-
"mean_acc_norm_ci":
|
| 344 |
-
"mean_acc": 0.
|
| 345 |
-
"mean_acc_ci":
|
| 346 |
-
"mean_centered_acc": -0.
|
| 347 |
-
"mean_centered_acc_ci":
|
| 348 |
"mean_acc_norm_runs": [
|
| 349 |
-
0.
|
| 350 |
-
0.
|
| 351 |
-
0.
|
| 352 |
-
0.
|
| 353 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 354 |
],
|
| 355 |
"mean_acc_runs": [
|
| 356 |
-
0.
|
| 357 |
-
0.
|
| 358 |
-
0.
|
| 359 |
-
0.
|
| 360 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 361 |
],
|
| 362 |
"mean_centered_acc_runs": [
|
| 363 |
-
|
| 364 |
-
|
| 365 |
-
|
| 366 |
-
|
| 367 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 368 |
],
|
| 369 |
-
"ci":
|
| 370 |
},
|
| 371 |
"batch_size_per_device": 8,
|
| 372 |
"device": "cuda:0",
|
|
@@ -376,4 +558,4 @@
|
|
| 376 |
"tokenizer_has_l2r_pred": true,
|
| 377 |
"tokenizer_has_next_1_pred": true,
|
| 378 |
"max_next_i_train": 5
|
| 379 |
-
}
|
|
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
|
|
|
|
|
|
|
|
|
| 12 |
"arc_challenge",
|
| 13 |
+
"arc_easy",
|
|
|
|
| 14 |
"boolq",
|
| 15 |
+
"commonsense_qa",
|
| 16 |
"copa",
|
| 17 |
+
"hellaswag",
|
| 18 |
"lambada_openai",
|
| 19 |
+
"openbookqa",
|
| 20 |
+
"piqa",
|
| 21 |
+
"rte",
|
| 22 |
+
"sciq",
|
| 23 |
+
"winogrande"
|
| 24 |
],
|
| 25 |
"tasks": {
|
| 26 |
"hellaswag": {
|
|
|
|
| 338 |
-0.08933333333333333
|
| 339 |
],
|
| 340 |
"ci": NaN
|
| 341 |
+
},
|
| 342 |
+
"rte": {
|
| 343 |
+
"acc_norm": null,
|
| 344 |
+
"acc_norm_ci": 0.0,
|
| 345 |
+
"acc_norm_runs": [],
|
| 346 |
+
"acc": 0.5270758122743683,
|
| 347 |
+
"acc_ci": NaN,
|
| 348 |
+
"acc_runs": [
|
| 349 |
+
0.5270758122743683,
|
| 350 |
+
0.5270758122743683,
|
| 351 |
+
0.5270758122743683,
|
| 352 |
+
0.5270758122743683,
|
| 353 |
+
0.5270758122743683
|
| 354 |
+
],
|
| 355 |
+
"perplexity": null,
|
| 356 |
+
"perplexity_ci": 0.0,
|
| 357 |
+
"perplexity_runs": [],
|
| 358 |
+
"stderr": 0.030052303463143738,
|
| 359 |
+
"centered_acc": 0.054151624548736566,
|
| 360 |
+
"centered_acc_ci": NaN,
|
| 361 |
+
"centered_acc_runs": [
|
| 362 |
+
0.054151624548736566,
|
| 363 |
+
0.054151624548736566,
|
| 364 |
+
0.054151624548736566,
|
| 365 |
+
0.054151624548736566,
|
| 366 |
+
0.054151624548736566
|
| 367 |
+
],
|
| 368 |
+
"ci": NaN
|
| 369 |
+
},
|
| 370 |
+
"commonsense_qa": {
|
| 371 |
+
"acc_norm": null,
|
| 372 |
+
"acc_norm_ci": 0.0,
|
| 373 |
+
"acc_norm_runs": [],
|
| 374 |
+
"acc": 0.19574119574119575,
|
| 375 |
+
"acc_ci": NaN,
|
| 376 |
+
"acc_runs": [
|
| 377 |
+
0.19574119574119575,
|
| 378 |
+
0.19574119574119575,
|
| 379 |
+
0.19574119574119575,
|
| 380 |
+
0.19574119574119575,
|
| 381 |
+
0.19574119574119575
|
| 382 |
+
],
|
| 383 |
+
"perplexity": null,
|
| 384 |
+
"perplexity_ci": 0.0,
|
| 385 |
+
"perplexity_runs": [],
|
| 386 |
+
"stderr": 0.011359497363584397,
|
| 387 |
+
"centered_acc": null,
|
| 388 |
+
"centered_acc_ci": 0.0,
|
| 389 |
+
"centered_acc_runs": [],
|
| 390 |
+
"ci": NaN
|
| 391 |
}
|
| 392 |
},
|
| 393 |
"summary": {
|
| 394 |
+
"mean_acc_norm": 0.2989041360233149,
|
| 395 |
+
"mean_acc_norm_ci": 0.04021074265142155,
|
| 396 |
+
"mean_acc": 0.31963220043202195,
|
| 397 |
+
"mean_acc_ci": 0.04699814720972745,
|
| 398 |
+
"mean_centered_acc": -0.0014987768028504544,
|
| 399 |
+
"mean_centered_acc_ci": 0.026709096339947704,
|
| 400 |
"mean_acc_norm_runs": [
|
| 401 |
+
0.25871340370444135,
|
| 402 |
+
0.25871340370444135,
|
| 403 |
+
0.25871340370444135,
|
| 404 |
+
0.25871340370444135,
|
| 405 |
+
0.25871340370444135,
|
| 406 |
+
0.5233949945593036,
|
| 407 |
+
0.5233949945593036,
|
| 408 |
+
0.5233949945593036,
|
| 409 |
+
0.5233949945593036,
|
| 410 |
+
0.5233949945593036,
|
| 411 |
+
0.26557239057239057,
|
| 412 |
+
0.26557239057239057,
|
| 413 |
+
0.26557239057239057,
|
| 414 |
+
0.26557239057239057,
|
| 415 |
+
0.26557239057239057,
|
| 416 |
+
0.27474402730375425,
|
| 417 |
+
0.27474402730375425,
|
| 418 |
+
0.27474402730375425,
|
| 419 |
+
0.27474402730375425,
|
| 420 |
+
0.27474402730375425,
|
| 421 |
+
0.288,
|
| 422 |
+
0.288,
|
| 423 |
+
0.288,
|
| 424 |
+
0.288,
|
| 425 |
+
0.288,
|
| 426 |
+
0.183,
|
| 427 |
+
0.183,
|
| 428 |
+
0.183,
|
| 429 |
+
0.183,
|
| 430 |
+
0.183
|
| 431 |
],
|
| 432 |
"mean_acc_runs": [
|
| 433 |
+
0.2557259510057757,
|
| 434 |
+
0.2557259510057757,
|
| 435 |
+
0.2557259510057757,
|
| 436 |
+
0.2557259510057757,
|
| 437 |
+
0.2557259510057757,
|
| 438 |
+
0.5424374319912949,
|
| 439 |
+
0.5424374319912949,
|
| 440 |
+
0.5424374319912949,
|
| 441 |
+
0.5424374319912949,
|
| 442 |
+
0.5424374319912949,
|
| 443 |
+
0.2647306397306397,
|
| 444 |
+
0.2647306397306397,
|
| 445 |
+
0.2647306397306397,
|
| 446 |
+
0.2647306397306397,
|
| 447 |
+
0.2647306397306397,
|
| 448 |
+
0.21075085324232082,
|
| 449 |
+
0.21075085324232082,
|
| 450 |
+
0.21075085324232082,
|
| 451 |
+
0.21075085324232082,
|
| 452 |
+
0.21075085324232082,
|
| 453 |
+
0.16,
|
| 454 |
+
0.16,
|
| 455 |
+
0.16,
|
| 456 |
+
0.16,
|
| 457 |
+
0.16,
|
| 458 |
+
0.500394632991318,
|
| 459 |
+
0.500394632991318,
|
| 460 |
+
0.500394632991318,
|
| 461 |
+
0.500394632991318,
|
| 462 |
+
0.500394632991318,
|
| 463 |
+
0.3782874617737003,
|
| 464 |
+
0.3782874617737003,
|
| 465 |
+
0.3782874617737003,
|
| 466 |
+
0.3782874617737003,
|
| 467 |
+
0.3782874617737003,
|
| 468 |
+
0.55,
|
| 469 |
+
0.55,
|
| 470 |
+
0.55,
|
| 471 |
+
0.55,
|
| 472 |
+
0.55,
|
| 473 |
+
0.0,
|
| 474 |
+
0.0,
|
| 475 |
+
0.0,
|
| 476 |
+
0.0,
|
| 477 |
+
0.0,
|
| 478 |
+
0.046,
|
| 479 |
+
0.046,
|
| 480 |
+
0.046,
|
| 481 |
+
0.046,
|
| 482 |
+
0.046,
|
| 483 |
+
0.5270758122743683,
|
| 484 |
+
0.5270758122743683,
|
| 485 |
+
0.5270758122743683,
|
| 486 |
+
0.5270758122743683,
|
| 487 |
+
0.5270758122743683,
|
| 488 |
+
0.19574119574119575,
|
| 489 |
+
0.19574119574119575,
|
| 490 |
+
0.19574119574119575,
|
| 491 |
+
0.19574119574119575,
|
| 492 |
+
0.19574119574119575,
|
| 493 |
+
0.5240746268656716,
|
| 494 |
+
0.5240746268656716,
|
| 495 |
+
0.5240746268656716,
|
| 496 |
+
0.5240746268656716,
|
| 497 |
+
0.5240746268656716
|
| 498 |
],
|
| 499 |
"mean_centered_acc_runs": [
|
| 500 |
+
0.011617871605921795,
|
| 501 |
+
0.011617871605921795,
|
| 502 |
+
0.011617871605921795,
|
| 503 |
+
0.011617871605921795,
|
| 504 |
+
0.011617871605921795,
|
| 505 |
+
0.04678998911860721,
|
| 506 |
+
0.04678998911860721,
|
| 507 |
+
0.04678998911860721,
|
| 508 |
+
0.04678998911860721,
|
| 509 |
+
0.04678998911860721,
|
| 510 |
+
0.020763187429854096,
|
| 511 |
+
0.020763187429854096,
|
| 512 |
+
0.020763187429854096,
|
| 513 |
+
0.020763187429854096,
|
| 514 |
+
0.020763187429854096,
|
| 515 |
+
0.03299203640500566,
|
| 516 |
+
0.03299203640500566,
|
| 517 |
+
0.03299203640500566,
|
| 518 |
+
0.03299203640500566,
|
| 519 |
+
0.03299203640500566,
|
| 520 |
+
0.05066666666666664,
|
| 521 |
+
0.05066666666666664,
|
| 522 |
+
0.05066666666666664,
|
| 523 |
+
0.05066666666666664,
|
| 524 |
+
0.05066666666666664,
|
| 525 |
+
0.0007892659826360671,
|
| 526 |
+
0.0007892659826360671,
|
| 527 |
+
0.0007892659826360671,
|
| 528 |
+
0.0007892659826360671,
|
| 529 |
+
0.0007892659826360671,
|
| 530 |
+
-0.24342507645259936,
|
| 531 |
+
-0.24342507645259936,
|
| 532 |
+
-0.24342507645259936,
|
| 533 |
+
-0.24342507645259936,
|
| 534 |
+
-0.24342507645259936,
|
| 535 |
+
0.10000000000000009,
|
| 536 |
+
0.10000000000000009,
|
| 537 |
+
0.10000000000000009,
|
| 538 |
+
0.10000000000000009,
|
| 539 |
+
0.10000000000000009,
|
| 540 |
+
-0.08933333333333333,
|
| 541 |
+
-0.08933333333333333,
|
| 542 |
+
-0.08933333333333333,
|
| 543 |
+
-0.08933333333333333,
|
| 544 |
+
-0.08933333333333333,
|
| 545 |
+
0.054151624548736566,
|
| 546 |
+
0.054151624548736566,
|
| 547 |
+
0.054151624548736566,
|
| 548 |
+
0.054151624548736566,
|
| 549 |
+
0.054151624548736566
|
| 550 |
],
|
| 551 |
+
"ci": 0.04021074265142155
|
| 552 |
},
|
| 553 |
"batch_size_per_device": 8,
|
| 554 |
"device": "cuda:0",
|
|
|
|
| 558 |
"tokenizer_has_l2r_pred": true,
|
| 559 |
"tokenizer_has_next_1_pred": true,
|
| 560 |
"max_next_i_train": 5
|
| 561 |
+
}
|
results/l2r50-i5-exp-fulle-lm-wsd-from-3168/zero_shot_eval/summary.jsonl
CHANGED
|
@@ -1 +1,3 @@
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/l2r50-i5-exp-fulle-lm-wsd-from-3168", "run": "l2r50-i5-exp-fulle-lm-wsd-from-3168", "checkpoint_path": "/workspace/concept_model/runs/l2r50-i5-exp-fulle-lm-wsd-from-3168/checkpoint-3312", "checkpoint": "checkpoint-3312", "global_step": 3312, "tokens_seen": 3472883712, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.25871340370444135, "acc_norm_ci": NaN, "acc_norm_runs": [0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135], "acc": 0.2557259510057757, "acc_ci": NaN, "acc_runs": [0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004370328224831643, "centered_acc": 0.011617871605921795, "centered_acc_ci": NaN, "centered_acc_runs": [0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795], "ci": NaN}, "piqa": {"acc_norm": 0.5233949945593036, "acc_norm_ci": NaN, "acc_norm_runs": [0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036], "acc": 0.5424374319912949, "acc_ci": NaN, "acc_runs": [0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.01165304715592769, "centered_acc": 0.04678998911860721, "centered_acc_ci": NaN, "centered_acc_runs": [0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721], "ci": NaN}, "arc_easy": {"acc_norm": 0.26557239057239057, "acc_norm_ci": NaN, "acc_norm_runs": [0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057], "acc": 0.2647306397306397, "acc_ci": NaN, "acc_runs": [0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00906221062697191, "centered_acc": 0.020763187429854096, "centered_acc_ci": NaN, "centered_acc_runs": [0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096], "ci": NaN}, "arc_challenge": {"acc_norm": 0.27474402730375425, "acc_norm_ci": NaN, "acc_norm_runs": [0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425], "acc": 0.21075085324232085, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.013044617212771207, "centered_acc": 0.03299203640500566, "centered_acc_ci": NaN, "centered_acc_runs": [0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566], "ci": NaN}, "openbookqa": {"acc_norm": 0.288, "acc_norm_ci": NaN, "acc_norm_runs": [0.288, 0.288, 0.288, 0.288, 0.288], "acc": 0.16, "acc_ci": NaN, "acc_runs": [0.16, 0.16, 0.16, 0.16, 0.16], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020271503835075255, "centered_acc": 0.05066666666666664, "centered_acc_ci": NaN, "centered_acc_runs": [0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.500394632991318, "acc_ci": NaN, "acc_runs": [0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052481306049611, "centered_acc": 0.0007892659826360671, "centered_acc_ci": NaN, "centered_acc_runs": [0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.55, "acc_ci": NaN, "acc_runs": [0.55, 0.55, 0.55, 0.55, 0.55], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05, "centered_acc": 0.10000000000000009, "centered_acc_ci": NaN, "centered_acc_runs": [0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 44957591.83115987, "perplexity_ci": NaN, "perplexity_runs": [44957591.83115987, 44957591.83115987, 44957591.83115987, 44957591.83115987, 44957591.83115987], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.183, "acc_norm_ci": NaN, "acc_norm_runs": [0.183, 0.183, 0.183, 0.183, 0.183], "acc": 0.046, "acc_ci": NaN, "acc_runs": [0.046, 0.046, 0.046, 0.046, 0.046], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012233587399477745, "centered_acc": -0.08933333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333], "ci": NaN}}, "summary": {"mean_acc_norm": 0.29890413602331495, "mean_acc_norm_ci": NaN, "mean_acc": 0.29083269707350495, "mean_acc_ci": NaN, "mean_centered_acc": -0.0076821547308045714, "mean_centered_acc_ci": 8.673617379884035e-19, "mean_acc_norm_runs": [0.29890413602331495, 0.29890413602331495, 0.29890413602331495, 0.29890413602331495, 0.29890413602331495], "mean_acc_runs": [0.29083269707350495, 0.29083269707350495, 0.29083269707350495, 0.29083269707350495, 0.29083269707350495], "mean_centered_acc_runs": [-0.007682154730804572, -0.007682154730804572, -0.007682154730804572, -0.007682154730804572, -0.007682154730804572], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
|
|
|
|
|
|
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/l2r50-i5-exp-fulle-lm-wsd-from-3168", "run": "l2r50-i5-exp-fulle-lm-wsd-from-3168", "checkpoint_path": "/workspace/concept_model/runs/l2r50-i5-exp-fulle-lm-wsd-from-3168/checkpoint-3312", "checkpoint": "checkpoint-3312", "global_step": 3312, "tokens_seen": 3472883712, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.25871340370444135, "acc_norm_ci": NaN, "acc_norm_runs": [0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135], "acc": 0.2557259510057757, "acc_ci": NaN, "acc_runs": [0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004370328224831643, "centered_acc": 0.011617871605921795, "centered_acc_ci": NaN, "centered_acc_runs": [0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795], "ci": NaN}, "piqa": {"acc_norm": 0.5233949945593036, "acc_norm_ci": NaN, "acc_norm_runs": [0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036], "acc": 0.5424374319912949, "acc_ci": NaN, "acc_runs": [0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.01165304715592769, "centered_acc": 0.04678998911860721, "centered_acc_ci": NaN, "centered_acc_runs": [0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721], "ci": NaN}, "arc_easy": {"acc_norm": 0.26557239057239057, "acc_norm_ci": NaN, "acc_norm_runs": [0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057], "acc": 0.2647306397306397, "acc_ci": NaN, "acc_runs": [0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00906221062697191, "centered_acc": 0.020763187429854096, "centered_acc_ci": NaN, "centered_acc_runs": [0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096], "ci": NaN}, "arc_challenge": {"acc_norm": 0.27474402730375425, "acc_norm_ci": NaN, "acc_norm_runs": [0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425], "acc": 0.21075085324232085, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.013044617212771207, "centered_acc": 0.03299203640500566, "centered_acc_ci": NaN, "centered_acc_runs": [0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566], "ci": NaN}, "openbookqa": {"acc_norm": 0.288, "acc_norm_ci": NaN, "acc_norm_runs": [0.288, 0.288, 0.288, 0.288, 0.288], "acc": 0.16, "acc_ci": NaN, "acc_runs": [0.16, 0.16, 0.16, 0.16, 0.16], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020271503835075255, "centered_acc": 0.05066666666666664, "centered_acc_ci": NaN, "centered_acc_runs": [0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.500394632991318, "acc_ci": NaN, "acc_runs": [0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052481306049611, "centered_acc": 0.0007892659826360671, "centered_acc_ci": NaN, "centered_acc_runs": [0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.55, "acc_ci": NaN, "acc_runs": [0.55, 0.55, 0.55, 0.55, 0.55], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05, "centered_acc": 0.10000000000000009, "centered_acc_ci": NaN, "centered_acc_runs": [0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 44957591.83115987, "perplexity_ci": NaN, "perplexity_runs": [44957591.83115987, 44957591.83115987, 44957591.83115987, 44957591.83115987, 44957591.83115987], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.183, "acc_norm_ci": NaN, "acc_norm_runs": [0.183, 0.183, 0.183, 0.183, 0.183], "acc": 0.046, "acc_ci": NaN, "acc_runs": [0.046, 0.046, 0.046, 0.046, 0.046], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012233587399477745, "centered_acc": -0.08933333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333], "ci": NaN}}, "summary": {"mean_acc_norm": 0.29890413602331495, "mean_acc_norm_ci": NaN, "mean_acc": 0.29083269707350495, "mean_acc_ci": NaN, "mean_centered_acc": -0.0076821547308045714, "mean_centered_acc_ci": 8.673617379884035e-19, "mean_acc_norm_runs": [0.29890413602331495, 0.29890413602331495, 0.29890413602331495, 0.29890413602331495, 0.29890413602331495], "mean_acc_runs": [0.29083269707350495, 0.29083269707350495, 0.29083269707350495, 0.29083269707350495, 0.29083269707350495], "mean_centered_acc_runs": [-0.007682154730804572, -0.007682154730804572, -0.007682154730804572, -0.007682154730804572, -0.007682154730804572], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
| 2 |
+
{"run_dir": "/workspace/concept_model/runs/l2r50-i5-exp-fulle-lm-wsd-from-3168", "run": "l2r50-i5-exp-fulle-lm-wsd-from-3168", "checkpoint_path": "/workspace/concept_model/runs/l2r50-i5-exp-fulle-lm-wsd-from-3168/checkpoint-3312", "checkpoint": "checkpoint-3312", "global_step": 3312, "tokens_seen": 3472883712, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.25871340370444135, "acc_norm_ci": NaN, "acc_norm_runs": [0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135], "acc": 0.2557259510057757, "acc_ci": NaN, "acc_runs": [0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004370328224831643, "centered_acc": 0.011617871605921795, "centered_acc_ci": NaN, "centered_acc_runs": [0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795], "ci": NaN}, "piqa": {"acc_norm": 0.5233949945593036, "acc_norm_ci": NaN, "acc_norm_runs": [0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036], "acc": 0.5424374319912949, "acc_ci": NaN, "acc_runs": [0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.01165304715592769, "centered_acc": 0.04678998911860721, "centered_acc_ci": NaN, "centered_acc_runs": [0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721], "ci": NaN}, "arc_easy": {"acc_norm": 0.26557239057239057, "acc_norm_ci": NaN, "acc_norm_runs": [0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057], "acc": 0.2647306397306397, "acc_ci": NaN, "acc_runs": [0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00906221062697191, "centered_acc": 0.020763187429854096, "centered_acc_ci": NaN, "centered_acc_runs": [0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096], "ci": NaN}, "arc_challenge": {"acc_norm": 0.27474402730375425, "acc_norm_ci": NaN, "acc_norm_runs": [0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425], "acc": 0.21075085324232085, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.013044617212771207, "centered_acc": 0.03299203640500566, "centered_acc_ci": NaN, "centered_acc_runs": [0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566], "ci": NaN}, "openbookqa": {"acc_norm": 0.288, "acc_norm_ci": NaN, "acc_norm_runs": [0.288, 0.288, 0.288, 0.288, 0.288], "acc": 0.16, "acc_ci": NaN, "acc_runs": [0.16, 0.16, 0.16, 0.16, 0.16], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020271503835075255, "centered_acc": 0.05066666666666664, "centered_acc_ci": NaN, "centered_acc_runs": [0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.500394632991318, "acc_ci": NaN, "acc_runs": [0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052481306049611, "centered_acc": 0.0007892659826360671, "centered_acc_ci": NaN, "centered_acc_runs": [0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.55, "acc_ci": NaN, "acc_runs": [0.55, 0.55, 0.55, 0.55, 0.55], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05, "centered_acc": 0.10000000000000009, "centered_acc_ci": NaN, "centered_acc_runs": [0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 44957591.83115987, "perplexity_ci": NaN, "perplexity_runs": [44957591.83115987, 44957591.83115987, 44957591.83115987, 44957591.83115987, 44957591.83115987], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.183, "acc_norm_ci": NaN, "acc_norm_runs": [0.183, 0.183, 0.183, 0.183, 0.183], "acc": 0.046, "acc_ci": NaN, "acc_runs": [0.046, 0.046, 0.046, 0.046, 0.046], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012233587399477745, "centered_acc": -0.08933333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.2989041360233149, "mean_acc_norm_ci": 0.04021074265142155, "mean_acc": 0.3025953315625511, "mean_acc_ci": 0.048457229783720523, "mean_centered_acc": -0.0014987768028504544, "mean_centered_acc_ci": 0.026709096339947704, "mean_acc_norm_runs": [0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.288, 0.288, 0.288, 0.288, 0.288, 0.183, 0.183, 0.183, 0.183, 0.183], "mean_acc_runs": [0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.16, 0.16, 0.16, 0.16, 0.16, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.55, 0.55, 0.55, 0.55, 0.55, 0.0, 0.0, 0.0, 0.0, 0.0, 0.046, 0.046, 0.046, 0.046, 0.046, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "mean_centered_acc_runs": [0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.04021074265142155}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
| 3 |
+
{"run_dir": "/workspace/concept_model/runs/l2r50-i5-exp-fulle-lm-wsd-from-3168", "run": "l2r50-i5-exp-fulle-lm-wsd-from-3168", "checkpoint_path": "/workspace/concept_model/runs/l2r50-i5-exp-fulle-lm-wsd-from-3168/checkpoint-3312", "checkpoint": "checkpoint-3312", "global_step": 3312, "tokens_seen": 3472883712, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "blimp", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.25871340370444135, "acc_norm_ci": NaN, "acc_norm_runs": [0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135], "acc": 0.2557259510057757, "acc_ci": NaN, "acc_runs": [0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004370328224831643, "centered_acc": 0.011617871605921795, "centered_acc_ci": NaN, "centered_acc_runs": [0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795], "ci": NaN}, "piqa": {"acc_norm": 0.5233949945593036, "acc_norm_ci": NaN, "acc_norm_runs": [0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036], "acc": 0.5424374319912949, "acc_ci": NaN, "acc_runs": [0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.01165304715592769, "centered_acc": 0.04678998911860721, "centered_acc_ci": NaN, "centered_acc_runs": [0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721], "ci": NaN}, "arc_easy": {"acc_norm": 0.26557239057239057, "acc_norm_ci": NaN, "acc_norm_runs": [0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057], "acc": 0.2647306397306397, "acc_ci": NaN, "acc_runs": [0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00906221062697191, "centered_acc": 0.020763187429854096, "centered_acc_ci": NaN, "centered_acc_runs": [0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096], "ci": NaN}, "arc_challenge": {"acc_norm": 0.27474402730375425, "acc_norm_ci": NaN, "acc_norm_runs": [0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425], "acc": 0.21075085324232085, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.013044617212771207, "centered_acc": 0.03299203640500566, "centered_acc_ci": NaN, "centered_acc_runs": [0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566], "ci": NaN}, "openbookqa": {"acc_norm": 0.288, "acc_norm_ci": NaN, "acc_norm_runs": [0.288, 0.288, 0.288, 0.288, 0.288], "acc": 0.16, "acc_ci": NaN, "acc_runs": [0.16, 0.16, 0.16, 0.16, 0.16], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020271503835075255, "centered_acc": 0.05066666666666664, "centered_acc_ci": NaN, "centered_acc_runs": [0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.500394632991318, "acc_ci": NaN, "acc_runs": [0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052481306049611, "centered_acc": 0.0007892659826360671, "centered_acc_ci": NaN, "centered_acc_runs": [0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.55, "acc_ci": NaN, "acc_runs": [0.55, 0.55, 0.55, 0.55, 0.55], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05, "centered_acc": 0.10000000000000009, "centered_acc_ci": NaN, "centered_acc_runs": [0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 44957591.83115987, "perplexity_ci": NaN, "perplexity_runs": [44957591.83115987, 44957591.83115987, 44957591.83115987, 44957591.83115987, 44957591.83115987], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.183, "acc_norm_ci": NaN, "acc_norm_runs": [0.183, 0.183, 0.183, 0.183, 0.183], "acc": 0.046, "acc_ci": NaN, "acc_runs": [0.046, 0.046, 0.046, 0.046, 0.046], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012233587399477745, "centered_acc": -0.08933333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "blimp": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5240746268656716, "acc_ci": NaN, "acc_runs": [0.5240746268656716, 0.5240746268656716, 0.5240746268656716, 0.5240746268656716, 0.5240746268656716], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0018234141012600888, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.2989041360233149, "mean_acc_norm_ci": 0.04021074265142155, "mean_acc": 0.31963220043202195, "mean_acc_ci": 0.04699814720972745, "mean_centered_acc": -0.0014987768028504544, "mean_centered_acc_ci": 0.026709096339947704, "mean_acc_norm_runs": [0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.25871340370444135, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.5233949945593036, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.26557239057239057, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.27474402730375425, 0.288, 0.288, 0.288, 0.288, 0.288, 0.183, 0.183, 0.183, 0.183, 0.183], "mean_acc_runs": [0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.2557259510057757, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.5424374319912949, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.2647306397306397, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.21075085324232082, 0.16, 0.16, 0.16, 0.16, 0.16, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.55, 0.55, 0.55, 0.55, 0.55, 0.0, 0.0, 0.0, 0.0, 0.0, 0.046, 0.046, 0.046, 0.046, 0.046, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.5240746268656716, 0.5240746268656716, 0.5240746268656716, 0.5240746268656716, 0.5240746268656716], "mean_centered_acc_runs": [0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.011617871605921795, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.04678998911860721, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.020763187429854096, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.03299203640500566, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.05066666666666664, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009, 0.10000000000000009, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, -0.08933333333333333, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.04021074265142155}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
results/random15-fulle-lm-wsd-from-2016/zero_shot_eval/checkpoint-2088.json
CHANGED
|
@@ -9,16 +9,18 @@
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
| 12 |
-
"hellaswag",
|
| 13 |
-
"piqa",
|
| 14 |
-
"arc_easy",
|
| 15 |
"arc_challenge",
|
| 16 |
-
"
|
| 17 |
-
"winogrande",
|
| 18 |
"boolq",
|
|
|
|
| 19 |
"copa",
|
|
|
|
| 20 |
"lambada_openai",
|
| 21 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
],
|
| 23 |
"tasks": {
|
| 24 |
"hellaswag": {
|
|
@@ -336,37 +338,217 @@
|
|
| 336 |
-0.072
|
| 337 |
],
|
| 338 |
"ci": NaN
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 339 |
}
|
| 340 |
},
|
| 341 |
"summary": {
|
| 342 |
-
"mean_acc_norm": 0.
|
| 343 |
-
"mean_acc_norm_ci":
|
| 344 |
-
"mean_acc": 0.
|
| 345 |
-
"mean_acc_ci":
|
| 346 |
-
"mean_centered_acc": 0.
|
| 347 |
-
"mean_centered_acc_ci":
|
| 348 |
"mean_acc_norm_runs": [
|
| 349 |
-
0.
|
| 350 |
-
0.
|
| 351 |
-
0.
|
| 352 |
-
0.
|
| 353 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 354 |
],
|
| 355 |
"mean_acc_runs": [
|
| 356 |
-
0.
|
| 357 |
-
0.
|
| 358 |
-
0.
|
| 359 |
-
0.
|
| 360 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 361 |
],
|
| 362 |
"mean_centered_acc_runs": [
|
| 363 |
-
0.
|
| 364 |
-
0.
|
| 365 |
-
0.
|
| 366 |
-
0.
|
| 367 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 368 |
],
|
| 369 |
-
"ci":
|
| 370 |
},
|
| 371 |
"batch_size_per_device": 8,
|
| 372 |
"device": "cuda:0",
|
|
@@ -376,4 +558,4 @@
|
|
| 376 |
"tokenizer_has_l2r_pred": true,
|
| 377 |
"tokenizer_has_next_1_pred": false,
|
| 378 |
"max_next_i_train": 1
|
| 379 |
-
}
|
|
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
|
|
|
|
|
|
|
|
|
| 12 |
"arc_challenge",
|
| 13 |
+
"arc_easy",
|
|
|
|
| 14 |
"boolq",
|
| 15 |
+
"commonsense_qa",
|
| 16 |
"copa",
|
| 17 |
+
"hellaswag",
|
| 18 |
"lambada_openai",
|
| 19 |
+
"openbookqa",
|
| 20 |
+
"piqa",
|
| 21 |
+
"rte",
|
| 22 |
+
"sciq",
|
| 23 |
+
"winogrande"
|
| 24 |
],
|
| 25 |
"tasks": {
|
| 26 |
"hellaswag": {
|
|
|
|
| 338 |
-0.072
|
| 339 |
],
|
| 340 |
"ci": NaN
|
| 341 |
+
},
|
| 342 |
+
"rte": {
|
| 343 |
+
"acc_norm": null,
|
| 344 |
+
"acc_norm_ci": 0.0,
|
| 345 |
+
"acc_norm_runs": [],
|
| 346 |
+
"acc": 0.5270758122743683,
|
| 347 |
+
"acc_ci": NaN,
|
| 348 |
+
"acc_runs": [
|
| 349 |
+
0.5270758122743683,
|
| 350 |
+
0.5270758122743683,
|
| 351 |
+
0.5270758122743683,
|
| 352 |
+
0.5270758122743683,
|
| 353 |
+
0.5270758122743683
|
| 354 |
+
],
|
| 355 |
+
"perplexity": null,
|
| 356 |
+
"perplexity_ci": 0.0,
|
| 357 |
+
"perplexity_runs": [],
|
| 358 |
+
"stderr": 0.030052303463143738,
|
| 359 |
+
"centered_acc": 0.054151624548736566,
|
| 360 |
+
"centered_acc_ci": NaN,
|
| 361 |
+
"centered_acc_runs": [
|
| 362 |
+
0.054151624548736566,
|
| 363 |
+
0.054151624548736566,
|
| 364 |
+
0.054151624548736566,
|
| 365 |
+
0.054151624548736566,
|
| 366 |
+
0.054151624548736566
|
| 367 |
+
],
|
| 368 |
+
"ci": NaN
|
| 369 |
+
},
|
| 370 |
+
"commonsense_qa": {
|
| 371 |
+
"acc_norm": null,
|
| 372 |
+
"acc_norm_ci": 0.0,
|
| 373 |
+
"acc_norm_runs": [],
|
| 374 |
+
"acc": 0.19574119574119575,
|
| 375 |
+
"acc_ci": NaN,
|
| 376 |
+
"acc_runs": [
|
| 377 |
+
0.19574119574119575,
|
| 378 |
+
0.19574119574119575,
|
| 379 |
+
0.19574119574119575,
|
| 380 |
+
0.19574119574119575,
|
| 381 |
+
0.19574119574119575
|
| 382 |
+
],
|
| 383 |
+
"perplexity": null,
|
| 384 |
+
"perplexity_ci": 0.0,
|
| 385 |
+
"perplexity_runs": [],
|
| 386 |
+
"stderr": 0.011359497363584397,
|
| 387 |
+
"centered_acc": null,
|
| 388 |
+
"centered_acc_ci": 0.0,
|
| 389 |
+
"centered_acc_runs": [],
|
| 390 |
+
"ci": NaN
|
| 391 |
}
|
| 392 |
},
|
| 393 |
"summary": {
|
| 394 |
+
"mean_acc_norm": 0.3111555362974754,
|
| 395 |
+
"mean_acc_norm_ci": 0.04676905483730215,
|
| 396 |
+
"mean_acc": 0.3337011783122285,
|
| 397 |
+
"mean_acc_ci": 0.05290039851946993,
|
| 398 |
+
"mean_centered_acc": 0.009942242314855035,
|
| 399 |
+
"mean_centered_acc_ci": 0.029224609002263327,
|
| 400 |
"mean_acc_norm_runs": [
|
| 401 |
+
0.2546305516829317,
|
| 402 |
+
0.2546305516829317,
|
| 403 |
+
0.2546305516829317,
|
| 404 |
+
0.2546305516829317,
|
| 405 |
+
0.2546305516829317,
|
| 406 |
+
0.5756256800870512,
|
| 407 |
+
0.5756256800870512,
|
| 408 |
+
0.5756256800870512,
|
| 409 |
+
0.5756256800870512,
|
| 410 |
+
0.5756256800870512,
|
| 411 |
+
0.27441077441077444,
|
| 412 |
+
0.27441077441077444,
|
| 413 |
+
0.27441077441077444,
|
| 414 |
+
0.27441077441077444,
|
| 415 |
+
0.27441077441077444,
|
| 416 |
+
0.25426621160409557,
|
| 417 |
+
0.25426621160409557,
|
| 418 |
+
0.25426621160409557,
|
| 419 |
+
0.25426621160409557,
|
| 420 |
+
0.25426621160409557,
|
| 421 |
+
0.312,
|
| 422 |
+
0.312,
|
| 423 |
+
0.312,
|
| 424 |
+
0.312,
|
| 425 |
+
0.312,
|
| 426 |
+
0.196,
|
| 427 |
+
0.196,
|
| 428 |
+
0.196,
|
| 429 |
+
0.196,
|
| 430 |
+
0.196
|
| 431 |
],
|
| 432 |
"mean_acc_runs": [
|
| 433 |
+
0.26249751045608444,
|
| 434 |
+
0.26249751045608444,
|
| 435 |
+
0.26249751045608444,
|
| 436 |
+
0.26249751045608444,
|
| 437 |
+
0.26249751045608444,
|
| 438 |
+
0.5903155603917302,
|
| 439 |
+
0.5903155603917302,
|
| 440 |
+
0.5903155603917302,
|
| 441 |
+
0.5903155603917302,
|
| 442 |
+
0.5903155603917302,
|
| 443 |
+
0.26936026936026936,
|
| 444 |
+
0.26936026936026936,
|
| 445 |
+
0.26936026936026936,
|
| 446 |
+
0.26936026936026936,
|
| 447 |
+
0.26936026936026936,
|
| 448 |
+
0.18771331058020477,
|
| 449 |
+
0.18771331058020477,
|
| 450 |
+
0.18771331058020477,
|
| 451 |
+
0.18771331058020477,
|
| 452 |
+
0.18771331058020477,
|
| 453 |
+
0.14,
|
| 454 |
+
0.14,
|
| 455 |
+
0.14,
|
| 456 |
+
0.14,
|
| 457 |
+
0.14,
|
| 458 |
+
0.5011838989739542,
|
| 459 |
+
0.5011838989739542,
|
| 460 |
+
0.5011838989739542,
|
| 461 |
+
0.5011838989739542,
|
| 462 |
+
0.5011838989739542,
|
| 463 |
+
0.3782874617737003,
|
| 464 |
+
0.3782874617737003,
|
| 465 |
+
0.3782874617737003,
|
| 466 |
+
0.3782874617737003,
|
| 467 |
+
0.3782874617737003,
|
| 468 |
+
0.54,
|
| 469 |
+
0.54,
|
| 470 |
+
0.54,
|
| 471 |
+
0.54,
|
| 472 |
+
0.54,
|
| 473 |
+
0.0,
|
| 474 |
+
0.0,
|
| 475 |
+
0.0,
|
| 476 |
+
0.0,
|
| 477 |
+
0.0,
|
| 478 |
+
0.052,
|
| 479 |
+
0.052,
|
| 480 |
+
0.052,
|
| 481 |
+
0.052,
|
| 482 |
+
0.052,
|
| 483 |
+
0.5270758122743683,
|
| 484 |
+
0.5270758122743683,
|
| 485 |
+
0.5270758122743683,
|
| 486 |
+
0.5270758122743683,
|
| 487 |
+
0.5270758122743683,
|
| 488 |
+
0.19574119574119575,
|
| 489 |
+
0.19574119574119575,
|
| 490 |
+
0.19574119574119575,
|
| 491 |
+
0.19574119574119575,
|
| 492 |
+
0.19574119574119575,
|
| 493 |
+
0.6939402985074626,
|
| 494 |
+
0.6939402985074626,
|
| 495 |
+
0.6939402985074626,
|
| 496 |
+
0.6939402985074626,
|
| 497 |
+
0.6939402985074626
|
| 498 |
],
|
| 499 |
"mean_centered_acc_runs": [
|
| 500 |
+
0.006174068910575592,
|
| 501 |
+
0.006174068910575592,
|
| 502 |
+
0.006174068910575592,
|
| 503 |
+
0.006174068910575592,
|
| 504 |
+
0.006174068910575592,
|
| 505 |
+
0.15125136017410235,
|
| 506 |
+
0.15125136017410235,
|
| 507 |
+
0.15125136017410235,
|
| 508 |
+
0.15125136017410235,
|
| 509 |
+
0.15125136017410235,
|
| 510 |
+
0.03254769921436592,
|
| 511 |
+
0.03254769921436592,
|
| 512 |
+
0.03254769921436592,
|
| 513 |
+
0.03254769921436592,
|
| 514 |
+
0.03254769921436592,
|
| 515 |
+
0.0056882821387940945,
|
| 516 |
+
0.0056882821387940945,
|
| 517 |
+
0.0056882821387940945,
|
| 518 |
+
0.0056882821387940945,
|
| 519 |
+
0.0056882821387940945,
|
| 520 |
+
0.08266666666666667,
|
| 521 |
+
0.08266666666666667,
|
| 522 |
+
0.08266666666666667,
|
| 523 |
+
0.08266666666666667,
|
| 524 |
+
0.08266666666666667,
|
| 525 |
+
0.0023677979479084232,
|
| 526 |
+
0.0023677979479084232,
|
| 527 |
+
0.0023677979479084232,
|
| 528 |
+
0.0023677979479084232,
|
| 529 |
+
0.0023677979479084232,
|
| 530 |
+
-0.24342507645259936,
|
| 531 |
+
-0.24342507645259936,
|
| 532 |
+
-0.24342507645259936,
|
| 533 |
+
-0.24342507645259936,
|
| 534 |
+
-0.24342507645259936,
|
| 535 |
+
0.08000000000000007,
|
| 536 |
+
0.08000000000000007,
|
| 537 |
+
0.08000000000000007,
|
| 538 |
+
0.08000000000000007,
|
| 539 |
+
0.08000000000000007,
|
| 540 |
+
-0.072,
|
| 541 |
+
-0.072,
|
| 542 |
+
-0.072,
|
| 543 |
+
-0.072,
|
| 544 |
+
-0.072,
|
| 545 |
+
0.054151624548736566,
|
| 546 |
+
0.054151624548736566,
|
| 547 |
+
0.054151624548736566,
|
| 548 |
+
0.054151624548736566,
|
| 549 |
+
0.054151624548736566
|
| 550 |
],
|
| 551 |
+
"ci": 0.04676905483730215
|
| 552 |
},
|
| 553 |
"batch_size_per_device": 8,
|
| 554 |
"device": "cuda:0",
|
|
|
|
| 558 |
"tokenizer_has_l2r_pred": true,
|
| 559 |
"tokenizer_has_next_1_pred": false,
|
| 560 |
"max_next_i_train": 1
|
| 561 |
+
}
|
results/random15-fulle-lm-wsd-from-2016/zero_shot_eval/summary.jsonl
CHANGED
|
@@ -1 +1,3 @@
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/random15-fulle-lm-wsd-from-2016", "run": "random15-fulle-lm-wsd-from-2016", "checkpoint_path": "/workspace/concept_model/runs/random15-fulle-lm-wsd-from-2016/checkpoint-2088", "checkpoint": "checkpoint-2088", "global_step": 2088, "tokens_seen": 2189426688, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2546305516829317, "acc_norm_ci": NaN, "acc_norm_runs": [0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317], "acc": 0.26249751045608444, "acc_ci": NaN, "acc_runs": [0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004347629889041179, "centered_acc": 0.006174068910575592, "centered_acc_ci": NaN, "centered_acc_runs": [0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592], "ci": NaN}, "piqa": {"acc_norm": 0.5756256800870512, "acc_norm_ci": NaN, "acc_norm_runs": [0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512], "acc": 0.5903155603917302, "acc_ci": NaN, "acc_runs": [0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011531612758871207, "centered_acc": 0.15125136017410235, "centered_acc_ci": NaN, "centered_acc_runs": [0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235], "ci": NaN}, "arc_easy": {"acc_norm": 0.27441077441077444, "acc_norm_ci": NaN, "acc_norm_runs": [0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444], "acc": 0.26936026936026936, "acc_ci": NaN, "acc_runs": [0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0091561771222447, "centered_acc": 0.03254769921436592, "centered_acc_ci": NaN, "centered_acc_runs": [0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592], "ci": NaN}, "arc_challenge": {"acc_norm": 0.25426621160409557, "acc_norm_ci": NaN, "acc_norm_runs": [0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557], "acc": 0.18771331058020477, "acc_ci": NaN, "acc_runs": [0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012724999945157845, "centered_acc": 0.0056882821387940945, "centered_acc_ci": NaN, "centered_acc_runs": [0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945], "ci": NaN}, "openbookqa": {"acc_norm": 0.312, "acc_norm_ci": NaN, "acc_norm_runs": [0.312, 0.312, 0.312, 0.312, 0.312], "acc": 0.14, "acc_ci": NaN, "acc_runs": [0.14, 0.14, 0.14, 0.14, 0.14], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020740596536488017, "centered_acc": 0.08266666666666667, "centered_acc_ci": NaN, "centered_acc_runs": [0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5011838989739542, "acc_ci": NaN, "acc_runs": [0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052446290528871, "centered_acc": 0.0023677979479084232, "centered_acc_ci": NaN, "centered_acc_runs": [0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.54, "acc_ci": NaN, "acc_runs": [0.54, 0.54, 0.54, 0.54, 0.54], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05009082659620331, "centered_acc": 0.08000000000000007, "centered_acc_ci": NaN, "centered_acc_runs": [0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 528159.7559436635, "perplexity_ci": NaN, "perplexity_runs": [528159.7559436635, 528159.7559436635, 528159.7559436635, 528159.7559436635, 528159.7559436635], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.196, "acc_norm_ci": NaN, "acc_norm_runs": [0.196, 0.196, 0.196, 0.196, 0.196], "acc": 0.052000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.052, 0.052, 0.052, 0.052, 0.052], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012559527926707347, "centered_acc": -0.072, "centered_acc_ci": NaN, "centered_acc_runs": [-0.072, -0.072, -0.072, -0.072, -0.072], "ci": NaN}}, "summary": {"mean_acc_norm": 0.31115553629747544, "mean_acc_norm_ci": NaN, "mean_acc": 0.29213580115359433, "mean_acc_ci": NaN, "mean_centered_acc": 0.005030088733312641, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.31115553629747544, 0.31115553629747544, 0.31115553629747544, 0.31115553629747544, 0.31115553629747544], "mean_acc_runs": [0.29213580115359433, 0.29213580115359433, 0.29213580115359433, 0.29213580115359433, 0.29213580115359433], "mean_centered_acc_runs": [0.005030088733312641, 0.005030088733312641, 0.005030088733312641, 0.005030088733312641, 0.005030088733312641], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
|
|
|
|
|
|
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/random15-fulle-lm-wsd-from-2016", "run": "random15-fulle-lm-wsd-from-2016", "checkpoint_path": "/workspace/concept_model/runs/random15-fulle-lm-wsd-from-2016/checkpoint-2088", "checkpoint": "checkpoint-2088", "global_step": 2088, "tokens_seen": 2189426688, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2546305516829317, "acc_norm_ci": NaN, "acc_norm_runs": [0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317], "acc": 0.26249751045608444, "acc_ci": NaN, "acc_runs": [0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004347629889041179, "centered_acc": 0.006174068910575592, "centered_acc_ci": NaN, "centered_acc_runs": [0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592], "ci": NaN}, "piqa": {"acc_norm": 0.5756256800870512, "acc_norm_ci": NaN, "acc_norm_runs": [0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512], "acc": 0.5903155603917302, "acc_ci": NaN, "acc_runs": [0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011531612758871207, "centered_acc": 0.15125136017410235, "centered_acc_ci": NaN, "centered_acc_runs": [0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235], "ci": NaN}, "arc_easy": {"acc_norm": 0.27441077441077444, "acc_norm_ci": NaN, "acc_norm_runs": [0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444], "acc": 0.26936026936026936, "acc_ci": NaN, "acc_runs": [0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0091561771222447, "centered_acc": 0.03254769921436592, "centered_acc_ci": NaN, "centered_acc_runs": [0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592], "ci": NaN}, "arc_challenge": {"acc_norm": 0.25426621160409557, "acc_norm_ci": NaN, "acc_norm_runs": [0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557], "acc": 0.18771331058020477, "acc_ci": NaN, "acc_runs": [0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012724999945157845, "centered_acc": 0.0056882821387940945, "centered_acc_ci": NaN, "centered_acc_runs": [0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945], "ci": NaN}, "openbookqa": {"acc_norm": 0.312, "acc_norm_ci": NaN, "acc_norm_runs": [0.312, 0.312, 0.312, 0.312, 0.312], "acc": 0.14, "acc_ci": NaN, "acc_runs": [0.14, 0.14, 0.14, 0.14, 0.14], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020740596536488017, "centered_acc": 0.08266666666666667, "centered_acc_ci": NaN, "centered_acc_runs": [0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5011838989739542, "acc_ci": NaN, "acc_runs": [0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052446290528871, "centered_acc": 0.0023677979479084232, "centered_acc_ci": NaN, "centered_acc_runs": [0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.54, "acc_ci": NaN, "acc_runs": [0.54, 0.54, 0.54, 0.54, 0.54], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05009082659620331, "centered_acc": 0.08000000000000007, "centered_acc_ci": NaN, "centered_acc_runs": [0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 528159.7559436635, "perplexity_ci": NaN, "perplexity_runs": [528159.7559436635, 528159.7559436635, 528159.7559436635, 528159.7559436635, 528159.7559436635], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.196, "acc_norm_ci": NaN, "acc_norm_runs": [0.196, 0.196, 0.196, 0.196, 0.196], "acc": 0.052000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.052, 0.052, 0.052, 0.052, 0.052], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012559527926707347, "centered_acc": -0.072, "centered_acc_ci": NaN, "centered_acc_runs": [-0.072, -0.072, -0.072, -0.072, -0.072], "ci": NaN}}, "summary": {"mean_acc_norm": 0.31115553629747544, "mean_acc_norm_ci": NaN, "mean_acc": 0.29213580115359433, "mean_acc_ci": NaN, "mean_centered_acc": 0.005030088733312641, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.31115553629747544, 0.31115553629747544, 0.31115553629747544, 0.31115553629747544, 0.31115553629747544], "mean_acc_runs": [0.29213580115359433, 0.29213580115359433, 0.29213580115359433, 0.29213580115359433, 0.29213580115359433], "mean_centered_acc_runs": [0.005030088733312641, 0.005030088733312641, 0.005030088733312641, 0.005030088733312641, 0.005030088733312641], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
| 2 |
+
{"run_dir": "/workspace/concept_model/runs/random15-fulle-lm-wsd-from-2016", "run": "random15-fulle-lm-wsd-from-2016", "checkpoint_path": "/workspace/concept_model/runs/random15-fulle-lm-wsd-from-2016/checkpoint-2088", "checkpoint": "checkpoint-2088", "global_step": 2088, "tokens_seen": 2189426688, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2546305516829317, "acc_norm_ci": NaN, "acc_norm_runs": [0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317], "acc": 0.26249751045608444, "acc_ci": NaN, "acc_runs": [0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004347629889041179, "centered_acc": 0.006174068910575592, "centered_acc_ci": NaN, "centered_acc_runs": [0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592], "ci": NaN}, "piqa": {"acc_norm": 0.5756256800870512, "acc_norm_ci": NaN, "acc_norm_runs": [0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512], "acc": 0.5903155603917302, "acc_ci": NaN, "acc_runs": [0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011531612758871207, "centered_acc": 0.15125136017410235, "centered_acc_ci": NaN, "centered_acc_runs": [0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235], "ci": NaN}, "arc_easy": {"acc_norm": 0.27441077441077444, "acc_norm_ci": NaN, "acc_norm_runs": [0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444], "acc": 0.26936026936026936, "acc_ci": NaN, "acc_runs": [0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0091561771222447, "centered_acc": 0.03254769921436592, "centered_acc_ci": NaN, "centered_acc_runs": [0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592], "ci": NaN}, "arc_challenge": {"acc_norm": 0.25426621160409557, "acc_norm_ci": NaN, "acc_norm_runs": [0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557], "acc": 0.18771331058020477, "acc_ci": NaN, "acc_runs": [0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012724999945157845, "centered_acc": 0.0056882821387940945, "centered_acc_ci": NaN, "centered_acc_runs": [0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945], "ci": NaN}, "openbookqa": {"acc_norm": 0.312, "acc_norm_ci": NaN, "acc_norm_runs": [0.312, 0.312, 0.312, 0.312, 0.312], "acc": 0.14, "acc_ci": NaN, "acc_runs": [0.14, 0.14, 0.14, 0.14, 0.14], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020740596536488017, "centered_acc": 0.08266666666666667, "centered_acc_ci": NaN, "centered_acc_runs": [0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5011838989739542, "acc_ci": NaN, "acc_runs": [0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052446290528871, "centered_acc": 0.0023677979479084232, "centered_acc_ci": NaN, "centered_acc_runs": [0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.54, "acc_ci": NaN, "acc_runs": [0.54, 0.54, 0.54, 0.54, 0.54], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05009082659620331, "centered_acc": 0.08000000000000007, "centered_acc_ci": NaN, "centered_acc_runs": [0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 528159.7559436635, "perplexity_ci": NaN, "perplexity_runs": [528159.7559436635, 528159.7559436635, 528159.7559436635, 528159.7559436635, 528159.7559436635], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.196, "acc_norm_ci": NaN, "acc_norm_runs": [0.196, 0.196, 0.196, 0.196, 0.196], "acc": 0.052000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.052, 0.052, 0.052, 0.052, 0.052], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012559527926707347, "centered_acc": -0.072, "centered_acc_ci": NaN, "centered_acc_runs": [-0.072, -0.072, -0.072, -0.072, -0.072], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.3111555362974754, "mean_acc_norm_ci": 0.04676905483730215, "mean_acc": 0.30368125162929227, "mean_acc_ci": 0.05004270067152489, "mean_centered_acc": 0.009942242314855035, "mean_centered_acc_ci": 0.029224609002263327, "mean_acc_norm_runs": [0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.312, 0.312, 0.312, 0.312, 0.312, 0.196, 0.196, 0.196, 0.196, 0.196], "mean_acc_runs": [0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.14, 0.14, 0.14, 0.14, 0.14, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.54, 0.54, 0.54, 0.54, 0.54, 0.0, 0.0, 0.0, 0.0, 0.0, 0.052, 0.052, 0.052, 0.052, 0.052, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "mean_centered_acc_runs": [0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007, -0.072, -0.072, -0.072, -0.072, -0.072, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.04676905483730215}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
| 3 |
+
{"run_dir": "/workspace/concept_model/runs/random15-fulle-lm-wsd-from-2016", "run": "random15-fulle-lm-wsd-from-2016", "checkpoint_path": "/workspace/concept_model/runs/random15-fulle-lm-wsd-from-2016/checkpoint-2088", "checkpoint": "checkpoint-2088", "global_step": 2088, "tokens_seen": 2189426688, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "blimp", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2546305516829317, "acc_norm_ci": NaN, "acc_norm_runs": [0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317], "acc": 0.26249751045608444, "acc_ci": NaN, "acc_runs": [0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004347629889041179, "centered_acc": 0.006174068910575592, "centered_acc_ci": NaN, "centered_acc_runs": [0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592], "ci": NaN}, "piqa": {"acc_norm": 0.5756256800870512, "acc_norm_ci": NaN, "acc_norm_runs": [0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512], "acc": 0.5903155603917302, "acc_ci": NaN, "acc_runs": [0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011531612758871207, "centered_acc": 0.15125136017410235, "centered_acc_ci": NaN, "centered_acc_runs": [0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235], "ci": NaN}, "arc_easy": {"acc_norm": 0.27441077441077444, "acc_norm_ci": NaN, "acc_norm_runs": [0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444], "acc": 0.26936026936026936, "acc_ci": NaN, "acc_runs": [0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0091561771222447, "centered_acc": 0.03254769921436592, "centered_acc_ci": NaN, "centered_acc_runs": [0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592], "ci": NaN}, "arc_challenge": {"acc_norm": 0.25426621160409557, "acc_norm_ci": NaN, "acc_norm_runs": [0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557], "acc": 0.18771331058020477, "acc_ci": NaN, "acc_runs": [0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012724999945157845, "centered_acc": 0.0056882821387940945, "centered_acc_ci": NaN, "centered_acc_runs": [0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945], "ci": NaN}, "openbookqa": {"acc_norm": 0.312, "acc_norm_ci": NaN, "acc_norm_runs": [0.312, 0.312, 0.312, 0.312, 0.312], "acc": 0.14, "acc_ci": NaN, "acc_runs": [0.14, 0.14, 0.14, 0.14, 0.14], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020740596536488017, "centered_acc": 0.08266666666666667, "centered_acc_ci": NaN, "centered_acc_runs": [0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5011838989739542, "acc_ci": NaN, "acc_runs": [0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052446290528871, "centered_acc": 0.0023677979479084232, "centered_acc_ci": NaN, "centered_acc_runs": [0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.54, "acc_ci": NaN, "acc_runs": [0.54, 0.54, 0.54, 0.54, 0.54], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05009082659620331, "centered_acc": 0.08000000000000007, "centered_acc_ci": NaN, "centered_acc_runs": [0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 528159.7559436635, "perplexity_ci": NaN, "perplexity_runs": [528159.7559436635, 528159.7559436635, 528159.7559436635, 528159.7559436635, 528159.7559436635], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.196, "acc_norm_ci": NaN, "acc_norm_runs": [0.196, 0.196, 0.196, 0.196, 0.196], "acc": 0.052000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.052, 0.052, 0.052, 0.052, 0.052], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012559527926707347, "centered_acc": -0.072, "centered_acc_ci": NaN, "centered_acc_runs": [-0.072, -0.072, -0.072, -0.072, -0.072], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "blimp": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.6939402985074626, "acc_ci": NaN, "acc_runs": [0.6939402985074626, 0.6939402985074626, 0.6939402985074626, 0.6939402985074626, 0.6939402985074626], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.001571782113767315, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.3111555362974754, "mean_acc_norm_ci": 0.04676905483730215, "mean_acc": 0.3337011783122285, "mean_acc_ci": 0.05290039851946993, "mean_centered_acc": 0.009942242314855035, "mean_centered_acc_ci": 0.029224609002263327, "mean_acc_norm_runs": [0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.2546305516829317, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.5756256800870512, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.27441077441077444, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.25426621160409557, 0.312, 0.312, 0.312, 0.312, 0.312, 0.196, 0.196, 0.196, 0.196, 0.196], "mean_acc_runs": [0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.26249751045608444, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.5903155603917302, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.26936026936026936, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.18771331058020477, 0.14, 0.14, 0.14, 0.14, 0.14, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.5011838989739542, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.54, 0.54, 0.54, 0.54, 0.54, 0.0, 0.0, 0.0, 0.0, 0.0, 0.052, 0.052, 0.052, 0.052, 0.052, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.6939402985074626, 0.6939402985074626, 0.6939402985074626, 0.6939402985074626, 0.6939402985074626], "mean_centered_acc_runs": [0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.006174068910575592, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.15125136017410235, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.03254769921436592, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.0056882821387940945, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.08266666666666667, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, 0.0023677979479084232, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007, 0.08000000000000007, -0.072, -0.072, -0.072, -0.072, -0.072, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.04676905483730215}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
results/random15-i5-exp-fulle-lm-wsd-from-3168/zero_shot_eval/checkpoint-3672.json
CHANGED
|
@@ -9,16 +9,18 @@
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
| 12 |
-
"hellaswag",
|
| 13 |
-
"piqa",
|
| 14 |
-
"arc_easy",
|
| 15 |
"arc_challenge",
|
| 16 |
-
"
|
| 17 |
-
"winogrande",
|
| 18 |
"boolq",
|
|
|
|
| 19 |
"copa",
|
|
|
|
| 20 |
"lambada_openai",
|
| 21 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
],
|
| 23 |
"tasks": {
|
| 24 |
"hellaswag": {
|
|
@@ -336,37 +338,217 @@
|
|
| 336 |
-0.08533333333333333
|
| 337 |
],
|
| 338 |
"ci": NaN
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 339 |
}
|
| 340 |
},
|
| 341 |
"summary": {
|
| 342 |
"mean_acc_norm": 0.29432321847181814,
|
| 343 |
-
"mean_acc_norm_ci":
|
| 344 |
-
"mean_acc": 0.
|
| 345 |
-
"mean_acc_ci":
|
| 346 |
-
"mean_centered_acc": -0.
|
| 347 |
-
"mean_centered_acc_ci":
|
| 348 |
"mean_acc_norm_runs": [
|
| 349 |
-
0.
|
| 350 |
-
0.
|
| 351 |
-
0.
|
| 352 |
-
0.
|
| 353 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 354 |
],
|
| 355 |
"mean_acc_runs": [
|
| 356 |
-
0.
|
| 357 |
-
0.
|
| 358 |
-
0.
|
| 359 |
-
0.
|
| 360 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 361 |
],
|
| 362 |
"mean_centered_acc_runs": [
|
| 363 |
-
|
| 364 |
-
|
| 365 |
-
|
| 366 |
-
|
| 367 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 368 |
],
|
| 369 |
-
"ci":
|
| 370 |
},
|
| 371 |
"batch_size_per_device": 8,
|
| 372 |
"device": "cuda:0",
|
|
@@ -376,4 +558,4 @@
|
|
| 376 |
"tokenizer_has_l2r_pred": true,
|
| 377 |
"tokenizer_has_next_1_pred": true,
|
| 378 |
"max_next_i_train": 5
|
| 379 |
-
}
|
|
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
|
|
|
|
|
|
|
|
|
| 12 |
"arc_challenge",
|
| 13 |
+
"arc_easy",
|
|
|
|
| 14 |
"boolq",
|
| 15 |
+
"commonsense_qa",
|
| 16 |
"copa",
|
| 17 |
+
"hellaswag",
|
| 18 |
"lambada_openai",
|
| 19 |
+
"openbookqa",
|
| 20 |
+
"piqa",
|
| 21 |
+
"rte",
|
| 22 |
+
"sciq",
|
| 23 |
+
"winogrande"
|
| 24 |
],
|
| 25 |
"tasks": {
|
| 26 |
"hellaswag": {
|
|
|
|
| 338 |
-0.08533333333333333
|
| 339 |
],
|
| 340 |
"ci": NaN
|
| 341 |
+
},
|
| 342 |
+
"rte": {
|
| 343 |
+
"acc_norm": null,
|
| 344 |
+
"acc_norm_ci": 0.0,
|
| 345 |
+
"acc_norm_runs": [],
|
| 346 |
+
"acc": 0.5270758122743683,
|
| 347 |
+
"acc_ci": NaN,
|
| 348 |
+
"acc_runs": [
|
| 349 |
+
0.5270758122743683,
|
| 350 |
+
0.5270758122743683,
|
| 351 |
+
0.5270758122743683,
|
| 352 |
+
0.5270758122743683,
|
| 353 |
+
0.5270758122743683
|
| 354 |
+
],
|
| 355 |
+
"perplexity": null,
|
| 356 |
+
"perplexity_ci": 0.0,
|
| 357 |
+
"perplexity_runs": [],
|
| 358 |
+
"stderr": 0.030052303463143738,
|
| 359 |
+
"centered_acc": 0.054151624548736566,
|
| 360 |
+
"centered_acc_ci": NaN,
|
| 361 |
+
"centered_acc_runs": [
|
| 362 |
+
0.054151624548736566,
|
| 363 |
+
0.054151624548736566,
|
| 364 |
+
0.054151624548736566,
|
| 365 |
+
0.054151624548736566,
|
| 366 |
+
0.054151624548736566
|
| 367 |
+
],
|
| 368 |
+
"ci": NaN
|
| 369 |
+
},
|
| 370 |
+
"commonsense_qa": {
|
| 371 |
+
"acc_norm": null,
|
| 372 |
+
"acc_norm_ci": 0.0,
|
| 373 |
+
"acc_norm_runs": [],
|
| 374 |
+
"acc": 0.19574119574119575,
|
| 375 |
+
"acc_ci": NaN,
|
| 376 |
+
"acc_runs": [
|
| 377 |
+
0.19574119574119575,
|
| 378 |
+
0.19574119574119575,
|
| 379 |
+
0.19574119574119575,
|
| 380 |
+
0.19574119574119575,
|
| 381 |
+
0.19574119574119575
|
| 382 |
+
],
|
| 383 |
+
"perplexity": null,
|
| 384 |
+
"perplexity_ci": 0.0,
|
| 385 |
+
"perplexity_runs": [],
|
| 386 |
+
"stderr": 0.011359497363584397,
|
| 387 |
+
"centered_acc": null,
|
| 388 |
+
"centered_acc_ci": 0.0,
|
| 389 |
+
"centered_acc_runs": [],
|
| 390 |
+
"ci": NaN
|
| 391 |
}
|
| 392 |
},
|
| 393 |
"summary": {
|
| 394 |
"mean_acc_norm": 0.29432321847181814,
|
| 395 |
+
"mean_acc_norm_ci": 0.041274472954479,
|
| 396 |
+
"mean_acc": 0.32215317724159653,
|
| 397 |
+
"mean_acc_ci": 0.04710993774522493,
|
| 398 |
+
"mean_centered_acc": -0.009452661600148684,
|
| 399 |
+
"mean_centered_acc_ci": 0.02502579688336765,
|
| 400 |
"mean_acc_norm_runs": [
|
| 401 |
+
0.2604062935670185,
|
| 402 |
+
0.2604062935670185,
|
| 403 |
+
0.2604062935670185,
|
| 404 |
+
0.2604062935670185,
|
| 405 |
+
0.2604062935670185,
|
| 406 |
+
0.5277475516866159,
|
| 407 |
+
0.5277475516866159,
|
| 408 |
+
0.5277475516866159,
|
| 409 |
+
0.5277475516866159,
|
| 410 |
+
0.5277475516866159,
|
| 411 |
+
0.27146464646464646,
|
| 412 |
+
0.27146464646464646,
|
| 413 |
+
0.27146464646464646,
|
| 414 |
+
0.27146464646464646,
|
| 415 |
+
0.27146464646464646,
|
| 416 |
+
0.24232081911262798,
|
| 417 |
+
0.24232081911262798,
|
| 418 |
+
0.24232081911262798,
|
| 419 |
+
0.24232081911262798,
|
| 420 |
+
0.24232081911262798,
|
| 421 |
+
0.278,
|
| 422 |
+
0.278,
|
| 423 |
+
0.278,
|
| 424 |
+
0.278,
|
| 425 |
+
0.278,
|
| 426 |
+
0.186,
|
| 427 |
+
0.186,
|
| 428 |
+
0.186,
|
| 429 |
+
0.186,
|
| 430 |
+
0.186
|
| 431 |
],
|
| 432 |
"mean_acc_runs": [
|
| 433 |
+
0.2591117307309301,
|
| 434 |
+
0.2591117307309301,
|
| 435 |
+
0.2591117307309301,
|
| 436 |
+
0.2591117307309301,
|
| 437 |
+
0.2591117307309301,
|
| 438 |
+
0.543525571273123,
|
| 439 |
+
0.543525571273123,
|
| 440 |
+
0.543525571273123,
|
| 441 |
+
0.543525571273123,
|
| 442 |
+
0.543525571273123,
|
| 443 |
+
0.265993265993266,
|
| 444 |
+
0.265993265993266,
|
| 445 |
+
0.265993265993266,
|
| 446 |
+
0.265993265993266,
|
| 447 |
+
0.265993265993266,
|
| 448 |
+
0.2030716723549488,
|
| 449 |
+
0.2030716723549488,
|
| 450 |
+
0.2030716723549488,
|
| 451 |
+
0.2030716723549488,
|
| 452 |
+
0.2030716723549488,
|
| 453 |
+
0.168,
|
| 454 |
+
0.168,
|
| 455 |
+
0.168,
|
| 456 |
+
0.168,
|
| 457 |
+
0.168,
|
| 458 |
+
0.5074980268350434,
|
| 459 |
+
0.5074980268350434,
|
| 460 |
+
0.5074980268350434,
|
| 461 |
+
0.5074980268350434,
|
| 462 |
+
0.5074980268350434,
|
| 463 |
+
0.3782874617737003,
|
| 464 |
+
0.3782874617737003,
|
| 465 |
+
0.3782874617737003,
|
| 466 |
+
0.3782874617737003,
|
| 467 |
+
0.3782874617737003,
|
| 468 |
+
0.52,
|
| 469 |
+
0.52,
|
| 470 |
+
0.52,
|
| 471 |
+
0.52,
|
| 472 |
+
0.52,
|
| 473 |
+
0.0,
|
| 474 |
+
0.0,
|
| 475 |
+
0.0,
|
| 476 |
+
0.0,
|
| 477 |
+
0.0,
|
| 478 |
+
0.055,
|
| 479 |
+
0.055,
|
| 480 |
+
0.055,
|
| 481 |
+
0.055,
|
| 482 |
+
0.055,
|
| 483 |
+
0.5270758122743683,
|
| 484 |
+
0.5270758122743683,
|
| 485 |
+
0.5270758122743683,
|
| 486 |
+
0.5270758122743683,
|
| 487 |
+
0.5270758122743683,
|
| 488 |
+
0.19574119574119575,
|
| 489 |
+
0.19574119574119575,
|
| 490 |
+
0.19574119574119575,
|
| 491 |
+
0.19574119574119575,
|
| 492 |
+
0.19574119574119575,
|
| 493 |
+
0.5646865671641792,
|
| 494 |
+
0.5646865671641792,
|
| 495 |
+
0.5646865671641792,
|
| 496 |
+
0.5646865671641792,
|
| 497 |
+
0.5646865671641792
|
| 498 |
],
|
| 499 |
"mean_centered_acc_runs": [
|
| 500 |
+
0.013875058089357998,
|
| 501 |
+
0.013875058089357998,
|
| 502 |
+
0.013875058089357998,
|
| 503 |
+
0.013875058089357998,
|
| 504 |
+
0.013875058089357998,
|
| 505 |
+
0.055495103373231824,
|
| 506 |
+
0.055495103373231824,
|
| 507 |
+
0.055495103373231824,
|
| 508 |
+
0.055495103373231824,
|
| 509 |
+
0.055495103373231824,
|
| 510 |
+
0.028619528619528618,
|
| 511 |
+
0.028619528619528618,
|
| 512 |
+
0.028619528619528618,
|
| 513 |
+
0.028619528619528618,
|
| 514 |
+
0.028619528619528618,
|
| 515 |
+
-0.010238907849829356,
|
| 516 |
+
-0.010238907849829356,
|
| 517 |
+
-0.010238907849829356,
|
| 518 |
+
-0.010238907849829356,
|
| 519 |
+
-0.010238907849829356,
|
| 520 |
+
0.037333333333333364,
|
| 521 |
+
0.037333333333333364,
|
| 522 |
+
0.037333333333333364,
|
| 523 |
+
0.037333333333333364,
|
| 524 |
+
0.037333333333333364,
|
| 525 |
+
0.014996053670086829,
|
| 526 |
+
0.014996053670086829,
|
| 527 |
+
0.014996053670086829,
|
| 528 |
+
0.014996053670086829,
|
| 529 |
+
0.014996053670086829,
|
| 530 |
+
-0.24342507645259936,
|
| 531 |
+
-0.24342507645259936,
|
| 532 |
+
-0.24342507645259936,
|
| 533 |
+
-0.24342507645259936,
|
| 534 |
+
-0.24342507645259936,
|
| 535 |
+
0.040000000000000036,
|
| 536 |
+
0.040000000000000036,
|
| 537 |
+
0.040000000000000036,
|
| 538 |
+
0.040000000000000036,
|
| 539 |
+
0.040000000000000036,
|
| 540 |
+
-0.08533333333333333,
|
| 541 |
+
-0.08533333333333333,
|
| 542 |
+
-0.08533333333333333,
|
| 543 |
+
-0.08533333333333333,
|
| 544 |
+
-0.08533333333333333,
|
| 545 |
+
0.054151624548736566,
|
| 546 |
+
0.054151624548736566,
|
| 547 |
+
0.054151624548736566,
|
| 548 |
+
0.054151624548736566,
|
| 549 |
+
0.054151624548736566
|
| 550 |
],
|
| 551 |
+
"ci": 0.041274472954479
|
| 552 |
},
|
| 553 |
"batch_size_per_device": 8,
|
| 554 |
"device": "cuda:0",
|
|
|
|
| 558 |
"tokenizer_has_l2r_pred": true,
|
| 559 |
"tokenizer_has_next_1_pred": true,
|
| 560 |
"max_next_i_train": 5
|
| 561 |
+
}
|
results/random15-i5-exp-fulle-lm-wsd-from-3168/zero_shot_eval/summary.jsonl
CHANGED
|
@@ -1 +1,3 @@
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/random15-i5-exp-fulle-lm-wsd-from-3168", "run": "random15-i5-exp-fulle-lm-wsd-from-3168", "checkpoint_path": "/workspace/concept_model/runs/random15-i5-exp-fulle-lm-wsd-from-3168/checkpoint-3672", "checkpoint": "checkpoint-3672", "global_step": 3672, "tokens_seen": 3850371072, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2604062935670185, "acc_norm_ci": NaN, "acc_norm_runs": [0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185], "acc": 0.2591117307309301, "acc_ci": NaN, "acc_runs": [0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004379594059141427, "centered_acc": 0.013875058089357998, "centered_acc_ci": NaN, "centered_acc_runs": [0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998], "ci": NaN}, "piqa": {"acc_norm": 0.5277475516866159, "acc_norm_ci": NaN, "acc_norm_runs": [0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159], "acc": 0.543525571273123, "acc_ci": NaN, "acc_runs": [0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011647846656062001, "centered_acc": 0.055495103373231824, "centered_acc_ci": NaN, "centered_acc_runs": [0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824], "ci": NaN}, "arc_easy": {"acc_norm": 0.27146464646464646, "acc_norm_ci": NaN, "acc_norm_runs": [0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646], "acc": 0.265993265993266, "acc_ci": NaN, "acc_runs": [0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00912536297036068, "centered_acc": 0.028619528619528618, "centered_acc_ci": NaN, "centered_acc_runs": [0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618], "ci": NaN}, "arc_challenge": {"acc_norm": 0.24232081911262798, "acc_norm_ci": NaN, "acc_norm_runs": [0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798], "acc": 0.20307167235494877, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012521593295800071, "centered_acc": -0.010238907849829356, "centered_acc_ci": NaN, "centered_acc_runs": [-0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356], "ci": NaN}, "openbookqa": {"acc_norm": 0.278, "acc_norm_ci": NaN, "acc_norm_runs": [0.278, 0.278, 0.278, 0.278, 0.278], "acc": 0.168, "acc_ci": NaN, "acc_runs": [0.168, 0.168, 0.168, 0.168, 0.168], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020055833888070865, "centered_acc": 0.037333333333333364, "centered_acc_ci": NaN, "centered_acc_runs": [0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5074980268350434, "acc_ci": NaN, "acc_runs": [0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014050905521228646, "centered_acc": 0.014996053670086829, "centered_acc_ci": NaN, "centered_acc_runs": [0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.52, "acc_ci": NaN, "acc_runs": [0.52, 0.52, 0.52, 0.52, 0.52], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05021167315686783, "centered_acc": 0.040000000000000036, "centered_acc_ci": NaN, "centered_acc_runs": [0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 4172100.838600532, "perplexity_ci": 4.656612873077393e-10, "perplexity_runs": [4172100.8386005317, 4172100.8386005317, 4172100.8386005317, 4172100.8386005317, 4172100.8386005317], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.186, "acc_norm_ci": NaN, "acc_norm_runs": [0.186, 0.186, 0.186, 0.186, 0.186], "acc": 0.05500000000000001, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.055, 0.055, 0.055, 0.055, 0.055], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012310790208412926, "centered_acc": -0.08533333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333], "ci": NaN}}, "summary": {"mean_acc_norm": 0.29432321847181814, "mean_acc_norm_ci": NaN, "mean_acc": 0.29004877289610115, "mean_acc_ci": NaN, "mean_centered_acc": -0.016519804505580375, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.29432321847181814, 0.29432321847181814, 0.29432321847181814, 0.29432321847181814, 0.29432321847181814], "mean_acc_runs": [0.29004877289610115, 0.29004877289610115, 0.29004877289610115, 0.29004877289610115, 0.29004877289610115], "mean_centered_acc_runs": [-0.016519804505580375, -0.016519804505580375, -0.016519804505580375, -0.016519804505580375, -0.016519804505580375], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
|
|
|
|
|
|
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/random15-i5-exp-fulle-lm-wsd-from-3168", "run": "random15-i5-exp-fulle-lm-wsd-from-3168", "checkpoint_path": "/workspace/concept_model/runs/random15-i5-exp-fulle-lm-wsd-from-3168/checkpoint-3672", "checkpoint": "checkpoint-3672", "global_step": 3672, "tokens_seen": 3850371072, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2604062935670185, "acc_norm_ci": NaN, "acc_norm_runs": [0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185], "acc": 0.2591117307309301, "acc_ci": NaN, "acc_runs": [0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004379594059141427, "centered_acc": 0.013875058089357998, "centered_acc_ci": NaN, "centered_acc_runs": [0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998], "ci": NaN}, "piqa": {"acc_norm": 0.5277475516866159, "acc_norm_ci": NaN, "acc_norm_runs": [0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159], "acc": 0.543525571273123, "acc_ci": NaN, "acc_runs": [0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011647846656062001, "centered_acc": 0.055495103373231824, "centered_acc_ci": NaN, "centered_acc_runs": [0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824], "ci": NaN}, "arc_easy": {"acc_norm": 0.27146464646464646, "acc_norm_ci": NaN, "acc_norm_runs": [0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646], "acc": 0.265993265993266, "acc_ci": NaN, "acc_runs": [0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00912536297036068, "centered_acc": 0.028619528619528618, "centered_acc_ci": NaN, "centered_acc_runs": [0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618], "ci": NaN}, "arc_challenge": {"acc_norm": 0.24232081911262798, "acc_norm_ci": NaN, "acc_norm_runs": [0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798], "acc": 0.20307167235494877, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012521593295800071, "centered_acc": -0.010238907849829356, "centered_acc_ci": NaN, "centered_acc_runs": [-0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356], "ci": NaN}, "openbookqa": {"acc_norm": 0.278, "acc_norm_ci": NaN, "acc_norm_runs": [0.278, 0.278, 0.278, 0.278, 0.278], "acc": 0.168, "acc_ci": NaN, "acc_runs": [0.168, 0.168, 0.168, 0.168, 0.168], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020055833888070865, "centered_acc": 0.037333333333333364, "centered_acc_ci": NaN, "centered_acc_runs": [0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5074980268350434, "acc_ci": NaN, "acc_runs": [0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014050905521228646, "centered_acc": 0.014996053670086829, "centered_acc_ci": NaN, "centered_acc_runs": [0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.52, "acc_ci": NaN, "acc_runs": [0.52, 0.52, 0.52, 0.52, 0.52], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05021167315686783, "centered_acc": 0.040000000000000036, "centered_acc_ci": NaN, "centered_acc_runs": [0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 4172100.838600532, "perplexity_ci": 4.656612873077393e-10, "perplexity_runs": [4172100.8386005317, 4172100.8386005317, 4172100.8386005317, 4172100.8386005317, 4172100.8386005317], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.186, "acc_norm_ci": NaN, "acc_norm_runs": [0.186, 0.186, 0.186, 0.186, 0.186], "acc": 0.05500000000000001, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.055, 0.055, 0.055, 0.055, 0.055], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012310790208412926, "centered_acc": -0.08533333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333], "ci": NaN}}, "summary": {"mean_acc_norm": 0.29432321847181814, "mean_acc_norm_ci": NaN, "mean_acc": 0.29004877289610115, "mean_acc_ci": NaN, "mean_centered_acc": -0.016519804505580375, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.29432321847181814, 0.29432321847181814, 0.29432321847181814, 0.29432321847181814, 0.29432321847181814], "mean_acc_runs": [0.29004877289610115, 0.29004877289610115, 0.29004877289610115, 0.29004877289610115, 0.29004877289610115], "mean_centered_acc_runs": [-0.016519804505580375, -0.016519804505580375, -0.016519804505580375, -0.016519804505580375, -0.016519804505580375], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
| 2 |
+
{"run_dir": "/workspace/concept_model/runs/random15-i5-exp-fulle-lm-wsd-from-3168", "run": "random15-i5-exp-fulle-lm-wsd-from-3168", "checkpoint_path": "/workspace/concept_model/runs/random15-i5-exp-fulle-lm-wsd-from-3168/checkpoint-3672", "checkpoint": "checkpoint-3672", "global_step": 3672, "tokens_seen": 3850371072, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2604062935670185, "acc_norm_ci": NaN, "acc_norm_runs": [0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185], "acc": 0.2591117307309301, "acc_ci": NaN, "acc_runs": [0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004379594059141427, "centered_acc": 0.013875058089357998, "centered_acc_ci": NaN, "centered_acc_runs": [0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998], "ci": NaN}, "piqa": {"acc_norm": 0.5277475516866159, "acc_norm_ci": NaN, "acc_norm_runs": [0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159], "acc": 0.543525571273123, "acc_ci": NaN, "acc_runs": [0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011647846656062001, "centered_acc": 0.055495103373231824, "centered_acc_ci": NaN, "centered_acc_runs": [0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824], "ci": NaN}, "arc_easy": {"acc_norm": 0.27146464646464646, "acc_norm_ci": NaN, "acc_norm_runs": [0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646], "acc": 0.265993265993266, "acc_ci": NaN, "acc_runs": [0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00912536297036068, "centered_acc": 0.028619528619528618, "centered_acc_ci": NaN, "centered_acc_runs": [0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618], "ci": NaN}, "arc_challenge": {"acc_norm": 0.24232081911262798, "acc_norm_ci": NaN, "acc_norm_runs": [0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798], "acc": 0.20307167235494877, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012521593295800071, "centered_acc": -0.010238907849829356, "centered_acc_ci": NaN, "centered_acc_runs": [-0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356], "ci": NaN}, "openbookqa": {"acc_norm": 0.278, "acc_norm_ci": NaN, "acc_norm_runs": [0.278, 0.278, 0.278, 0.278, 0.278], "acc": 0.168, "acc_ci": NaN, "acc_runs": [0.168, 0.168, 0.168, 0.168, 0.168], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020055833888070865, "centered_acc": 0.037333333333333364, "centered_acc_ci": NaN, "centered_acc_runs": [0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5074980268350434, "acc_ci": NaN, "acc_runs": [0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014050905521228646, "centered_acc": 0.014996053670086829, "centered_acc_ci": NaN, "centered_acc_runs": [0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.52, "acc_ci": NaN, "acc_runs": [0.52, 0.52, 0.52, 0.52, 0.52], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05021167315686783, "centered_acc": 0.040000000000000036, "centered_acc_ci": NaN, "centered_acc_runs": [0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 4172100.838600532, "perplexity_ci": 4.656612873077393e-10, "perplexity_runs": [4172100.8386005317, 4172100.8386005317, 4172100.8386005317, 4172100.8386005317, 4172100.8386005317], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.186, "acc_norm_ci": NaN, "acc_norm_runs": [0.186, 0.186, 0.186, 0.186, 0.186], "acc": 0.05500000000000001, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.055, 0.055, 0.055, 0.055, 0.055], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012310790208412926, "centered_acc": -0.08533333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.29432321847181814, "mean_acc_norm_ci": 0.041274472954479, "mean_acc": 0.30194206141471464, "mean_acc_ci": 0.04749943436443188, "mean_centered_acc": -0.009452661600148684, "mean_centered_acc_ci": 0.02502579688336765, "mean_acc_norm_runs": [0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.278, 0.278, 0.278, 0.278, 0.278, 0.186, 0.186, 0.186, 0.186, 0.186], "mean_acc_runs": [0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123, 0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.168, 0.168, 0.168, 0.168, 0.168, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.52, 0.52, 0.52, 0.52, 0.52, 0.0, 0.0, 0.0, 0.0, 0.0, 0.055, 0.055, 0.055, 0.055, 0.055, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "mean_centered_acc_runs": [0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.041274472954479}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
| 3 |
+
{"run_dir": "/workspace/concept_model/runs/random15-i5-exp-fulle-lm-wsd-from-3168", "run": "random15-i5-exp-fulle-lm-wsd-from-3168", "checkpoint_path": "/workspace/concept_model/runs/random15-i5-exp-fulle-lm-wsd-from-3168/checkpoint-3672", "checkpoint": "checkpoint-3672", "global_step": 3672, "tokens_seen": 3850371072, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "blimp", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2604062935670185, "acc_norm_ci": NaN, "acc_norm_runs": [0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185], "acc": 0.2591117307309301, "acc_ci": NaN, "acc_runs": [0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004379594059141427, "centered_acc": 0.013875058089357998, "centered_acc_ci": NaN, "centered_acc_runs": [0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998], "ci": NaN}, "piqa": {"acc_norm": 0.5277475516866159, "acc_norm_ci": NaN, "acc_norm_runs": [0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159], "acc": 0.543525571273123, "acc_ci": NaN, "acc_runs": [0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011647846656062001, "centered_acc": 0.055495103373231824, "centered_acc_ci": NaN, "centered_acc_runs": [0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824], "ci": NaN}, "arc_easy": {"acc_norm": 0.27146464646464646, "acc_norm_ci": NaN, "acc_norm_runs": [0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646], "acc": 0.265993265993266, "acc_ci": NaN, "acc_runs": [0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.00912536297036068, "centered_acc": 0.028619528619528618, "centered_acc_ci": NaN, "centered_acc_runs": [0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618], "ci": NaN}, "arc_challenge": {"acc_norm": 0.24232081911262798, "acc_norm_ci": NaN, "acc_norm_runs": [0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798], "acc": 0.20307167235494877, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012521593295800071, "centered_acc": -0.010238907849829356, "centered_acc_ci": NaN, "centered_acc_runs": [-0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356], "ci": NaN}, "openbookqa": {"acc_norm": 0.278, "acc_norm_ci": NaN, "acc_norm_runs": [0.278, 0.278, 0.278, 0.278, 0.278], "acc": 0.168, "acc_ci": NaN, "acc_runs": [0.168, 0.168, 0.168, 0.168, 0.168], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020055833888070865, "centered_acc": 0.037333333333333364, "centered_acc_ci": NaN, "centered_acc_runs": [0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5074980268350434, "acc_ci": NaN, "acc_runs": [0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014050905521228646, "centered_acc": 0.014996053670086829, "centered_acc_ci": NaN, "centered_acc_runs": [0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.52, "acc_ci": NaN, "acc_runs": [0.52, 0.52, 0.52, 0.52, 0.52], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.05021167315686783, "centered_acc": 0.040000000000000036, "centered_acc_ci": NaN, "centered_acc_runs": [0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 4172100.838600532, "perplexity_ci": 4.656612873077393e-10, "perplexity_runs": [4172100.8386005317, 4172100.8386005317, 4172100.8386005317, 4172100.8386005317, 4172100.8386005317], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.186, "acc_norm_ci": NaN, "acc_norm_runs": [0.186, 0.186, 0.186, 0.186, 0.186], "acc": 0.05500000000000001, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.055, 0.055, 0.055, 0.055, 0.055], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012310790208412926, "centered_acc": -0.08533333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "blimp": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5646865671641792, "acc_ci": NaN, "acc_runs": [0.5646865671641792, 0.5646865671641792, 0.5646865671641792, 0.5646865671641792, 0.5646865671641792], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0017976603302854194, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.29432321847181814, "mean_acc_norm_ci": 0.041274472954479, "mean_acc": 0.32215317724159653, "mean_acc_ci": 0.04710993774522493, "mean_centered_acc": -0.009452661600148684, "mean_centered_acc_ci": 0.02502579688336765, "mean_acc_norm_runs": [0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.2604062935670185, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.5277475516866159, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.27146464646464646, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.24232081911262798, 0.278, 0.278, 0.278, 0.278, 0.278, 0.186, 0.186, 0.186, 0.186, 0.186], "mean_acc_runs": [0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.2591117307309301, 0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123, 0.543525571273123, 0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266, 0.265993265993266, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.168, 0.168, 0.168, 0.168, 0.168, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.5074980268350434, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.52, 0.52, 0.52, 0.52, 0.52, 0.0, 0.0, 0.0, 0.0, 0.0, 0.055, 0.055, 0.055, 0.055, 0.055, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.5646865671641792, 0.5646865671641792, 0.5646865671641792, 0.5646865671641792, 0.5646865671641792], "mean_centered_acc_runs": [0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.013875058089357998, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.055495103373231824, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618, 0.028619528619528618, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356, -0.010238907849829356, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.037333333333333364, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829, 0.014996053670086829, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036, 0.040000000000000036, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.041274472954479}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": false, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
results/random15-l2r50-fulle-lm-wsd-from-7488/zero_shot_eval/checkpoint-7776.json
CHANGED
|
@@ -9,16 +9,18 @@
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
| 12 |
-
"hellaswag",
|
| 13 |
-
"piqa",
|
| 14 |
-
"arc_easy",
|
| 15 |
"arc_challenge",
|
| 16 |
-
"
|
| 17 |
-
"winogrande",
|
| 18 |
"boolq",
|
|
|
|
| 19 |
"copa",
|
|
|
|
| 20 |
"lambada_openai",
|
| 21 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
],
|
| 23 |
"tasks": {
|
| 24 |
"hellaswag": {
|
|
@@ -336,37 +338,217 @@
|
|
| 336 |
-0.08533333333333333
|
| 337 |
],
|
| 338 |
"ci": NaN
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 339 |
}
|
| 340 |
},
|
| 341 |
"summary": {
|
| 342 |
-
"mean_acc_norm": 0.
|
| 343 |
-
"mean_acc_norm_ci":
|
| 344 |
-
"mean_acc": 0.
|
| 345 |
-
"mean_acc_ci":
|
| 346 |
-
"mean_centered_acc": -0.
|
| 347 |
-
"mean_centered_acc_ci":
|
| 348 |
"mean_acc_norm_runs": [
|
| 349 |
-
0.
|
| 350 |
-
0.
|
| 351 |
-
0.
|
| 352 |
-
0.
|
| 353 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 354 |
],
|
| 355 |
"mean_acc_runs": [
|
| 356 |
-
0.
|
| 357 |
-
0.
|
| 358 |
-
0.
|
| 359 |
-
0.
|
| 360 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 361 |
],
|
| 362 |
"mean_centered_acc_runs": [
|
| 363 |
-
|
| 364 |
-
|
| 365 |
-
|
| 366 |
-
|
| 367 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 368 |
],
|
| 369 |
-
"ci":
|
| 370 |
},
|
| 371 |
"batch_size_per_device": 8,
|
| 372 |
"device": "cuda:0",
|
|
@@ -376,4 +558,4 @@
|
|
| 376 |
"tokenizer_has_l2r_pred": true,
|
| 377 |
"tokenizer_has_next_1_pred": false,
|
| 378 |
"max_next_i_train": 1
|
| 379 |
-
}
|
|
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
|
|
|
|
|
|
|
|
|
| 12 |
"arc_challenge",
|
| 13 |
+
"arc_easy",
|
|
|
|
| 14 |
"boolq",
|
| 15 |
+
"commonsense_qa",
|
| 16 |
"copa",
|
| 17 |
+
"hellaswag",
|
| 18 |
"lambada_openai",
|
| 19 |
+
"openbookqa",
|
| 20 |
+
"piqa",
|
| 21 |
+
"rte",
|
| 22 |
+
"sciq",
|
| 23 |
+
"winogrande"
|
| 24 |
],
|
| 25 |
"tasks": {
|
| 26 |
"hellaswag": {
|
|
|
|
| 338 |
-0.08533333333333333
|
| 339 |
],
|
| 340 |
"ci": NaN
|
| 341 |
+
},
|
| 342 |
+
"rte": {
|
| 343 |
+
"acc_norm": null,
|
| 344 |
+
"acc_norm_ci": 0.0,
|
| 345 |
+
"acc_norm_runs": [],
|
| 346 |
+
"acc": 0.5270758122743683,
|
| 347 |
+
"acc_ci": NaN,
|
| 348 |
+
"acc_runs": [
|
| 349 |
+
0.5270758122743683,
|
| 350 |
+
0.5270758122743683,
|
| 351 |
+
0.5270758122743683,
|
| 352 |
+
0.5270758122743683,
|
| 353 |
+
0.5270758122743683
|
| 354 |
+
],
|
| 355 |
+
"perplexity": null,
|
| 356 |
+
"perplexity_ci": 0.0,
|
| 357 |
+
"perplexity_runs": [],
|
| 358 |
+
"stderr": 0.030052303463143738,
|
| 359 |
+
"centered_acc": 0.054151624548736566,
|
| 360 |
+
"centered_acc_ci": NaN,
|
| 361 |
+
"centered_acc_runs": [
|
| 362 |
+
0.054151624548736566,
|
| 363 |
+
0.054151624548736566,
|
| 364 |
+
0.054151624548736566,
|
| 365 |
+
0.054151624548736566,
|
| 366 |
+
0.054151624548736566
|
| 367 |
+
],
|
| 368 |
+
"ci": NaN
|
| 369 |
+
},
|
| 370 |
+
"commonsense_qa": {
|
| 371 |
+
"acc_norm": null,
|
| 372 |
+
"acc_norm_ci": 0.0,
|
| 373 |
+
"acc_norm_runs": [],
|
| 374 |
+
"acc": 0.19574119574119575,
|
| 375 |
+
"acc_ci": NaN,
|
| 376 |
+
"acc_runs": [
|
| 377 |
+
0.19574119574119575,
|
| 378 |
+
0.19574119574119575,
|
| 379 |
+
0.19574119574119575,
|
| 380 |
+
0.19574119574119575,
|
| 381 |
+
0.19574119574119575
|
| 382 |
+
],
|
| 383 |
+
"perplexity": null,
|
| 384 |
+
"perplexity_ci": 0.0,
|
| 385 |
+
"perplexity_runs": [],
|
| 386 |
+
"stderr": 0.011359497363584397,
|
| 387 |
+
"centered_acc": null,
|
| 388 |
+
"centered_acc_ci": 0.0,
|
| 389 |
+
"centered_acc_runs": [],
|
| 390 |
+
"ci": NaN
|
| 391 |
}
|
| 392 |
},
|
| 393 |
"summary": {
|
| 394 |
+
"mean_acc_norm": 0.2996818134591798,
|
| 395 |
+
"mean_acc_norm_ci": 0.041411111903803144,
|
| 396 |
+
"mean_acc": 0.32199826141691723,
|
| 397 |
+
"mean_acc_ci": 0.04731466789331612,
|
| 398 |
+
"mean_centered_acc": -0.007491207675754334,
|
| 399 |
+
"mean_centered_acc_ci": 0.025464162772653627,
|
| 400 |
"mean_acc_norm_runs": [
|
| 401 |
+
0.2622983469428401,
|
| 402 |
+
0.2622983469428401,
|
| 403 |
+
0.2622983469428401,
|
| 404 |
+
0.2622983469428401,
|
| 405 |
+
0.2622983469428401,
|
| 406 |
+
0.5315560391730142,
|
| 407 |
+
0.5315560391730142,
|
| 408 |
+
0.5315560391730142,
|
| 409 |
+
0.5315560391730142,
|
| 410 |
+
0.5315560391730142,
|
| 411 |
+
0.2676767676767677,
|
| 412 |
+
0.2676767676767677,
|
| 413 |
+
0.2676767676767677,
|
| 414 |
+
0.2676767676767677,
|
| 415 |
+
0.2676767676767677,
|
| 416 |
+
0.2525597269624573,
|
| 417 |
+
0.2525597269624573,
|
| 418 |
+
0.2525597269624573,
|
| 419 |
+
0.2525597269624573,
|
| 420 |
+
0.2525597269624573,
|
| 421 |
+
0.298,
|
| 422 |
+
0.298,
|
| 423 |
+
0.298,
|
| 424 |
+
0.298,
|
| 425 |
+
0.298,
|
| 426 |
+
0.186,
|
| 427 |
+
0.186,
|
| 428 |
+
0.186,
|
| 429 |
+
0.186,
|
| 430 |
+
0.186
|
| 431 |
],
|
| 432 |
"mean_acc_runs": [
|
| 433 |
+
0.25692093208524197,
|
| 434 |
+
0.25692093208524197,
|
| 435 |
+
0.25692093208524197,
|
| 436 |
+
0.25692093208524197,
|
| 437 |
+
0.25692093208524197,
|
| 438 |
+
0.5429815016322089,
|
| 439 |
+
0.5429815016322089,
|
| 440 |
+
0.5429815016322089,
|
| 441 |
+
0.5429815016322089,
|
| 442 |
+
0.5429815016322089,
|
| 443 |
+
0.27314814814814814,
|
| 444 |
+
0.27314814814814814,
|
| 445 |
+
0.27314814814814814,
|
| 446 |
+
0.27314814814814814,
|
| 447 |
+
0.27314814814814814,
|
| 448 |
+
0.2030716723549488,
|
| 449 |
+
0.2030716723549488,
|
| 450 |
+
0.2030716723549488,
|
| 451 |
+
0.2030716723549488,
|
| 452 |
+
0.2030716723549488,
|
| 453 |
+
0.164,
|
| 454 |
+
0.164,
|
| 455 |
+
0.164,
|
| 456 |
+
0.164,
|
| 457 |
+
0.164,
|
| 458 |
+
0.5146014206787688,
|
| 459 |
+
0.5146014206787688,
|
| 460 |
+
0.5146014206787688,
|
| 461 |
+
0.5146014206787688,
|
| 462 |
+
0.5146014206787688,
|
| 463 |
+
0.3782874617737003,
|
| 464 |
+
0.3782874617737003,
|
| 465 |
+
0.3782874617737003,
|
| 466 |
+
0.3782874617737003,
|
| 467 |
+
0.3782874617737003,
|
| 468 |
+
0.5,
|
| 469 |
+
0.5,
|
| 470 |
+
0.5,
|
| 471 |
+
0.5,
|
| 472 |
+
0.5,
|
| 473 |
+
0.0,
|
| 474 |
+
0.0,
|
| 475 |
+
0.0,
|
| 476 |
+
0.0,
|
| 477 |
+
0.0,
|
| 478 |
+
0.052,
|
| 479 |
+
0.052,
|
| 480 |
+
0.052,
|
| 481 |
+
0.052,
|
| 482 |
+
0.052,
|
| 483 |
+
0.5270758122743683,
|
| 484 |
+
0.5270758122743683,
|
| 485 |
+
0.5270758122743683,
|
| 486 |
+
0.5270758122743683,
|
| 487 |
+
0.5270758122743683,
|
| 488 |
+
0.19574119574119575,
|
| 489 |
+
0.19574119574119575,
|
| 490 |
+
0.19574119574119575,
|
| 491 |
+
0.19574119574119575,
|
| 492 |
+
0.19574119574119575,
|
| 493 |
+
0.5781492537313433,
|
| 494 |
+
0.5781492537313433,
|
| 495 |
+
0.5781492537313433,
|
| 496 |
+
0.5781492537313433,
|
| 497 |
+
0.5781492537313433
|
| 498 |
],
|
| 499 |
"mean_centered_acc_runs": [
|
| 500 |
+
0.0163977959237868,
|
| 501 |
+
0.0163977959237868,
|
| 502 |
+
0.0163977959237868,
|
| 503 |
+
0.0163977959237868,
|
| 504 |
+
0.0163977959237868,
|
| 505 |
+
0.06311207834602839,
|
| 506 |
+
0.06311207834602839,
|
| 507 |
+
0.06311207834602839,
|
| 508 |
+
0.06311207834602839,
|
| 509 |
+
0.06311207834602839,
|
| 510 |
+
0.023569023569023573,
|
| 511 |
+
0.023569023569023573,
|
| 512 |
+
0.023569023569023573,
|
| 513 |
+
0.023569023569023573,
|
| 514 |
+
0.023569023569023573,
|
| 515 |
+
0.003412969283276427,
|
| 516 |
+
0.003412969283276427,
|
| 517 |
+
0.003412969283276427,
|
| 518 |
+
0.003412969283276427,
|
| 519 |
+
0.003412969283276427,
|
| 520 |
+
0.06399999999999999,
|
| 521 |
+
0.06399999999999999,
|
| 522 |
+
0.06399999999999999,
|
| 523 |
+
0.06399999999999999,
|
| 524 |
+
0.06399999999999999,
|
| 525 |
+
0.02920284135753759,
|
| 526 |
+
0.02920284135753759,
|
| 527 |
+
0.02920284135753759,
|
| 528 |
+
0.02920284135753759,
|
| 529 |
+
0.02920284135753759,
|
| 530 |
+
-0.24342507645259936,
|
| 531 |
+
-0.24342507645259936,
|
| 532 |
+
-0.24342507645259936,
|
| 533 |
+
-0.24342507645259936,
|
| 534 |
+
-0.24342507645259936,
|
| 535 |
+
0.0,
|
| 536 |
+
0.0,
|
| 537 |
+
0.0,
|
| 538 |
+
0.0,
|
| 539 |
+
0.0,
|
| 540 |
+
-0.08533333333333333,
|
| 541 |
+
-0.08533333333333333,
|
| 542 |
+
-0.08533333333333333,
|
| 543 |
+
-0.08533333333333333,
|
| 544 |
+
-0.08533333333333333,
|
| 545 |
+
0.054151624548736566,
|
| 546 |
+
0.054151624548736566,
|
| 547 |
+
0.054151624548736566,
|
| 548 |
+
0.054151624548736566,
|
| 549 |
+
0.054151624548736566
|
| 550 |
],
|
| 551 |
+
"ci": 0.041411111903803144
|
| 552 |
},
|
| 553 |
"batch_size_per_device": 8,
|
| 554 |
"device": "cuda:0",
|
|
|
|
| 558 |
"tokenizer_has_l2r_pred": true,
|
| 559 |
"tokenizer_has_next_1_pred": false,
|
| 560 |
"max_next_i_train": 1
|
| 561 |
+
}
|
results/random15-l2r50-fulle-lm-wsd-from-7488/zero_shot_eval/summary.jsonl
CHANGED
|
@@ -1 +1,3 @@
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/random15-l2r50-fulle-lm-wsd-from-7488", "run": "random15-l2r50-fulle-lm-wsd-from-7488", "checkpoint_path": "/workspace/concept_model/runs/random15-l2r50-fulle-lm-wsd-from-7488/checkpoint-7776", "checkpoint": "checkpoint-7776", "global_step": 7776, "tokens_seen": 8153726976, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2622983469428401, "acc_norm_ci": NaN, "acc_norm_runs": [0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401], "acc": 0.25692093208524197, "acc_ci": NaN, "acc_runs": [0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004389849907040485, "centered_acc": 0.0163977959237868, "centered_acc_ci": NaN, "centered_acc_runs": [0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868], "ci": NaN}, "piqa": {"acc_norm": 0.5315560391730142, "acc_norm_ci": NaN, "acc_norm_runs": [0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142], "acc": 0.5429815016322089, "acc_ci": NaN, "acc_runs": [0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011642567710878958, "centered_acc": 0.06311207834602839, "centered_acc_ci": NaN, "centered_acc_runs": [0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839], "ci": NaN}, "arc_easy": {"acc_norm": 0.2676767676767677, "acc_norm_ci": NaN, "acc_norm_runs": [0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677], "acc": 0.27314814814814814, "acc_ci": NaN, "acc_runs": [0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009085000147099275, "centered_acc": 0.023569023569023573, "centered_acc_ci": NaN, "centered_acc_runs": [0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573], "ci": NaN}, "arc_challenge": {"acc_norm": 0.2525597269624573, "acc_norm_ci": NaN, "acc_norm_runs": [0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573], "acc": 0.20307167235494877, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012696728980207687, "centered_acc": 0.003412969283276427, "centered_acc_ci": NaN, "centered_acc_runs": [0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427], "ci": NaN}, "openbookqa": {"acc_norm": 0.298, "acc_norm_ci": NaN, "acc_norm_runs": [0.298, 0.298, 0.298, 0.298, 0.298], "acc": 0.164, "acc_ci": NaN, "acc_runs": [0.164, 0.164, 0.164, 0.164, 0.164], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020475118092988954, "centered_acc": 0.06399999999999999, "centered_acc_ci": NaN, "centered_acc_runs": [0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5146014206787688, "acc_ci": NaN, "acc_runs": [0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014046492383275955, "centered_acc": 0.02920284135753759, "centered_acc_ci": NaN, "centered_acc_runs": [0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5, "acc_ci": NaN, "acc_runs": [0.5, 0.5, 0.5, 0.5, 0.5], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.050251890762960605, "centered_acc": 0.0, "centered_acc_ci": NaN, "centered_acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 5028151.863835744, "perplexity_ci": NaN, "perplexity_runs": [5028151.863835744, 5028151.863835744, 5028151.863835744, 5028151.863835744, 5028151.863835744], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.186, "acc_norm_ci": NaN, "acc_norm_runs": [0.186, 0.186, 0.186, 0.186, 0.186], "acc": 0.052000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.052, 0.052, 0.052, 0.052, 0.052], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012310790208412926, "centered_acc": -0.08533333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333], "ci": NaN}}, "summary": {"mean_acc_norm": 0.2996818134591799, "mean_acc_norm_ci": NaN, "mean_acc": 0.2885011136673017, "mean_acc_ci": NaN, "mean_centered_acc": -0.014340411256253325, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.2996818134591799, 0.2996818134591799, 0.2996818134591799, 0.2996818134591799, 0.2996818134591799], "mean_acc_runs": [0.2885011136673017, 0.2885011136673017, 0.2885011136673017, 0.2885011136673017, 0.2885011136673017], "mean_centered_acc_runs": [-0.014340411256253325, -0.014340411256253325, -0.014340411256253325, -0.014340411256253325, -0.014340411256253325], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
|
|
|
|
|
|
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/random15-l2r50-fulle-lm-wsd-from-7488", "run": "random15-l2r50-fulle-lm-wsd-from-7488", "checkpoint_path": "/workspace/concept_model/runs/random15-l2r50-fulle-lm-wsd-from-7488/checkpoint-7776", "checkpoint": "checkpoint-7776", "global_step": 7776, "tokens_seen": 8153726976, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2622983469428401, "acc_norm_ci": NaN, "acc_norm_runs": [0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401], "acc": 0.25692093208524197, "acc_ci": NaN, "acc_runs": [0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004389849907040485, "centered_acc": 0.0163977959237868, "centered_acc_ci": NaN, "centered_acc_runs": [0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868], "ci": NaN}, "piqa": {"acc_norm": 0.5315560391730142, "acc_norm_ci": NaN, "acc_norm_runs": [0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142], "acc": 0.5429815016322089, "acc_ci": NaN, "acc_runs": [0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011642567710878958, "centered_acc": 0.06311207834602839, "centered_acc_ci": NaN, "centered_acc_runs": [0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839], "ci": NaN}, "arc_easy": {"acc_norm": 0.2676767676767677, "acc_norm_ci": NaN, "acc_norm_runs": [0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677], "acc": 0.27314814814814814, "acc_ci": NaN, "acc_runs": [0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009085000147099275, "centered_acc": 0.023569023569023573, "centered_acc_ci": NaN, "centered_acc_runs": [0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573], "ci": NaN}, "arc_challenge": {"acc_norm": 0.2525597269624573, "acc_norm_ci": NaN, "acc_norm_runs": [0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573], "acc": 0.20307167235494877, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012696728980207687, "centered_acc": 0.003412969283276427, "centered_acc_ci": NaN, "centered_acc_runs": [0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427], "ci": NaN}, "openbookqa": {"acc_norm": 0.298, "acc_norm_ci": NaN, "acc_norm_runs": [0.298, 0.298, 0.298, 0.298, 0.298], "acc": 0.164, "acc_ci": NaN, "acc_runs": [0.164, 0.164, 0.164, 0.164, 0.164], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020475118092988954, "centered_acc": 0.06399999999999999, "centered_acc_ci": NaN, "centered_acc_runs": [0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5146014206787688, "acc_ci": NaN, "acc_runs": [0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014046492383275955, "centered_acc": 0.02920284135753759, "centered_acc_ci": NaN, "centered_acc_runs": [0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5, "acc_ci": NaN, "acc_runs": [0.5, 0.5, 0.5, 0.5, 0.5], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.050251890762960605, "centered_acc": 0.0, "centered_acc_ci": NaN, "centered_acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 5028151.863835744, "perplexity_ci": NaN, "perplexity_runs": [5028151.863835744, 5028151.863835744, 5028151.863835744, 5028151.863835744, 5028151.863835744], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.186, "acc_norm_ci": NaN, "acc_norm_runs": [0.186, 0.186, 0.186, 0.186, 0.186], "acc": 0.052000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.052, 0.052, 0.052, 0.052, 0.052], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012310790208412926, "centered_acc": -0.08533333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333], "ci": NaN}}, "summary": {"mean_acc_norm": 0.2996818134591799, "mean_acc_norm_ci": NaN, "mean_acc": 0.2885011136673017, "mean_acc_ci": NaN, "mean_centered_acc": -0.014340411256253325, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.2996818134591799, 0.2996818134591799, 0.2996818134591799, 0.2996818134591799, 0.2996818134591799], "mean_acc_runs": [0.2885011136673017, 0.2885011136673017, 0.2885011136673017, 0.2885011136673017, 0.2885011136673017], "mean_centered_acc_runs": [-0.014340411256253325, -0.014340411256253325, -0.014340411256253325, -0.014340411256253325, -0.014340411256253325], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
| 2 |
+
{"run_dir": "/workspace/concept_model/runs/random15-l2r50-fulle-lm-wsd-from-7488", "run": "random15-l2r50-fulle-lm-wsd-from-7488", "checkpoint_path": "/workspace/concept_model/runs/random15-l2r50-fulle-lm-wsd-from-7488/checkpoint-7776", "checkpoint": "checkpoint-7776", "global_step": 7776, "tokens_seen": 8153726976, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2622983469428401, "acc_norm_ci": NaN, "acc_norm_runs": [0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401], "acc": 0.25692093208524197, "acc_ci": NaN, "acc_runs": [0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004389849907040485, "centered_acc": 0.0163977959237868, "centered_acc_ci": NaN, "centered_acc_runs": [0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868], "ci": NaN}, "piqa": {"acc_norm": 0.5315560391730142, "acc_norm_ci": NaN, "acc_norm_runs": [0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142], "acc": 0.5429815016322089, "acc_ci": NaN, "acc_runs": [0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011642567710878958, "centered_acc": 0.06311207834602839, "centered_acc_ci": NaN, "centered_acc_runs": [0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839], "ci": NaN}, "arc_easy": {"acc_norm": 0.2676767676767677, "acc_norm_ci": NaN, "acc_norm_runs": [0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677], "acc": 0.27314814814814814, "acc_ci": NaN, "acc_runs": [0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009085000147099275, "centered_acc": 0.023569023569023573, "centered_acc_ci": NaN, "centered_acc_runs": [0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573], "ci": NaN}, "arc_challenge": {"acc_norm": 0.2525597269624573, "acc_norm_ci": NaN, "acc_norm_runs": [0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573], "acc": 0.20307167235494877, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012696728980207687, "centered_acc": 0.003412969283276427, "centered_acc_ci": NaN, "centered_acc_runs": [0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427], "ci": NaN}, "openbookqa": {"acc_norm": 0.298, "acc_norm_ci": NaN, "acc_norm_runs": [0.298, 0.298, 0.298, 0.298, 0.298], "acc": 0.164, "acc_ci": NaN, "acc_runs": [0.164, 0.164, 0.164, 0.164, 0.164], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020475118092988954, "centered_acc": 0.06399999999999999, "centered_acc_ci": NaN, "centered_acc_runs": [0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5146014206787688, "acc_ci": NaN, "acc_runs": [0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014046492383275955, "centered_acc": 0.02920284135753759, "centered_acc_ci": NaN, "centered_acc_runs": [0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5, "acc_ci": NaN, "acc_runs": [0.5, 0.5, 0.5, 0.5, 0.5], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.050251890762960605, "centered_acc": 0.0, "centered_acc_ci": NaN, "centered_acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 5028151.863835744, "perplexity_ci": NaN, "perplexity_runs": [5028151.863835744, 5028151.863835744, 5028151.863835744, 5028151.863835744, 5028151.863835744], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.186, "acc_norm_ci": NaN, "acc_norm_runs": [0.186, 0.186, 0.186, 0.186, 0.186], "acc": 0.052000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.052, 0.052, 0.052, 0.052, 0.052], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012310790208412926, "centered_acc": -0.08533333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.2996818134591798, "mean_acc_norm_ci": 0.041411111903803144, "mean_acc": 0.300652345390715, "mean_acc_ci": 0.047300980254449454, "mean_centered_acc": -0.007491207675754334, "mean_centered_acc_ci": 0.025464162772653627, "mean_acc_norm_runs": [0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.298, 0.298, 0.298, 0.298, 0.298, 0.186, 0.186, 0.186, 0.186, 0.186], "mean_acc_runs": [0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.164, 0.164, 0.164, 0.164, 0.164, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.5, 0.5, 0.5, 0.5, 0.5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.052, 0.052, 0.052, 0.052, 0.052, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "mean_centered_acc_runs": [0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.0, 0.0, 0.0, 0.0, 0.0, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.041411111903803144}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
| 3 |
+
{"run_dir": "/workspace/concept_model/runs/random15-l2r50-fulle-lm-wsd-from-7488", "run": "random15-l2r50-fulle-lm-wsd-from-7488", "checkpoint_path": "/workspace/concept_model/runs/random15-l2r50-fulle-lm-wsd-from-7488/checkpoint-7776", "checkpoint": "checkpoint-7776", "global_step": 7776, "tokens_seen": 8153726976, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "blimp", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2622983469428401, "acc_norm_ci": NaN, "acc_norm_runs": [0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401], "acc": 0.25692093208524197, "acc_ci": NaN, "acc_runs": [0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004389849907040485, "centered_acc": 0.0163977959237868, "centered_acc_ci": NaN, "centered_acc_runs": [0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868], "ci": NaN}, "piqa": {"acc_norm": 0.5315560391730142, "acc_norm_ci": NaN, "acc_norm_runs": [0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142], "acc": 0.5429815016322089, "acc_ci": NaN, "acc_runs": [0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011642567710878958, "centered_acc": 0.06311207834602839, "centered_acc_ci": NaN, "centered_acc_runs": [0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839], "ci": NaN}, "arc_easy": {"acc_norm": 0.2676767676767677, "acc_norm_ci": NaN, "acc_norm_runs": [0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677], "acc": 0.27314814814814814, "acc_ci": NaN, "acc_runs": [0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.009085000147099275, "centered_acc": 0.023569023569023573, "centered_acc_ci": NaN, "centered_acc_runs": [0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573], "ci": NaN}, "arc_challenge": {"acc_norm": 0.2525597269624573, "acc_norm_ci": NaN, "acc_norm_runs": [0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573], "acc": 0.20307167235494877, "acc_ci": 2.7755575615628914e-17, "acc_runs": [0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012696728980207687, "centered_acc": 0.003412969283276427, "centered_acc_ci": NaN, "centered_acc_runs": [0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427], "ci": NaN}, "openbookqa": {"acc_norm": 0.298, "acc_norm_ci": NaN, "acc_norm_runs": [0.298, 0.298, 0.298, 0.298, 0.298], "acc": 0.164, "acc_ci": NaN, "acc_runs": [0.164, 0.164, 0.164, 0.164, 0.164], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020475118092988954, "centered_acc": 0.06399999999999999, "centered_acc_ci": NaN, "centered_acc_runs": [0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5146014206787688, "acc_ci": NaN, "acc_runs": [0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014046492383275955, "centered_acc": 0.02920284135753759, "centered_acc_ci": NaN, "centered_acc_runs": [0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5, "acc_ci": NaN, "acc_runs": [0.5, 0.5, 0.5, 0.5, 0.5], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.050251890762960605, "centered_acc": 0.0, "centered_acc_ci": NaN, "centered_acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 5028151.863835744, "perplexity_ci": NaN, "perplexity_runs": [5028151.863835744, 5028151.863835744, 5028151.863835744, 5028151.863835744, 5028151.863835744], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.186, "acc_norm_ci": NaN, "acc_norm_runs": [0.186, 0.186, 0.186, 0.186, 0.186], "acc": 0.052000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.052, 0.052, 0.052, 0.052, 0.052], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012310790208412926, "centered_acc": -0.08533333333333333, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "blimp": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5781492537313433, "acc_ci": NaN, "acc_runs": [0.5781492537313433, 0.5781492537313433, 0.5781492537313433, 0.5781492537313433, 0.5781492537313433], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.0017752176705457604, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.2996818134591798, "mean_acc_norm_ci": 0.041411111903803144, "mean_acc": 0.32199826141691723, "mean_acc_ci": 0.04731466789331612, "mean_centered_acc": -0.007491207675754334, "mean_centered_acc_ci": 0.025464162772653627, "mean_acc_norm_runs": [0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.2622983469428401, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.5315560391730142, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2676767676767677, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.2525597269624573, 0.298, 0.298, 0.298, 0.298, 0.298, 0.186, 0.186, 0.186, 0.186, 0.186], "mean_acc_runs": [0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.25692093208524197, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.5429815016322089, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.27314814814814814, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.2030716723549488, 0.164, 0.164, 0.164, 0.164, 0.164, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.5146014206787688, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.5, 0.5, 0.5, 0.5, 0.5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.052, 0.052, 0.052, 0.052, 0.052, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.5781492537313433, 0.5781492537313433, 0.5781492537313433, 0.5781492537313433, 0.5781492537313433], "mean_centered_acc_runs": [0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.0163977959237868, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.06311207834602839, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.023569023569023573, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.003412969283276427, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.06399999999999999, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759, 0.02920284135753759, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.0, 0.0, 0.0, 0.0, 0.0, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, -0.08533333333333333, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.041411111903803144}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": false, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": false, "max_next_i_train": 1}
|
results/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024/zero_shot_eval/checkpoint-21960.json
CHANGED
|
@@ -9,16 +9,18 @@
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
| 12 |
-
"hellaswag",
|
| 13 |
-
"piqa",
|
| 14 |
-
"arc_easy",
|
| 15 |
"arc_challenge",
|
| 16 |
-
"
|
| 17 |
-
"winogrande",
|
| 18 |
"boolq",
|
|
|
|
| 19 |
"copa",
|
|
|
|
| 20 |
"lambada_openai",
|
| 21 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
],
|
| 23 |
"tasks": {
|
| 24 |
"hellaswag": {
|
|
@@ -336,37 +338,217 @@
|
|
| 336 |
-0.08
|
| 337 |
],
|
| 338 |
"ci": NaN
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 339 |
}
|
| 340 |
},
|
| 341 |
"summary": {
|
| 342 |
-
"mean_acc_norm": 0.
|
| 343 |
-
"mean_acc_norm_ci":
|
| 344 |
-
"mean_acc": 0.
|
| 345 |
-
"mean_acc_ci":
|
| 346 |
-
"mean_centered_acc": 0.
|
| 347 |
-
"mean_centered_acc_ci":
|
| 348 |
"mean_acc_norm_runs": [
|
| 349 |
-
0.
|
| 350 |
-
0.
|
| 351 |
-
0.
|
| 352 |
-
0.
|
| 353 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 354 |
],
|
| 355 |
"mean_acc_runs": [
|
| 356 |
-
0.
|
| 357 |
-
0.
|
| 358 |
-
0.
|
| 359 |
-
0.
|
| 360 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 361 |
],
|
| 362 |
"mean_centered_acc_runs": [
|
| 363 |
-
0.
|
| 364 |
-
0.
|
| 365 |
-
0.
|
| 366 |
-
0.
|
| 367 |
-
0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 368 |
],
|
| 369 |
-
"ci":
|
| 370 |
},
|
| 371 |
"batch_size_per_device": 8,
|
| 372 |
"device": "cuda:0",
|
|
@@ -376,4 +558,4 @@
|
|
| 376 |
"tokenizer_has_l2r_pred": true,
|
| 377 |
"tokenizer_has_next_1_pred": true,
|
| 378 |
"max_next_i_train": 5
|
| 379 |
-
}
|
|
|
|
| 9 |
"runs_per_task": 5,
|
| 10 |
"base_seed": 42,
|
| 11 |
"tasks_evaluated": [
|
|
|
|
|
|
|
|
|
|
| 12 |
"arc_challenge",
|
| 13 |
+
"arc_easy",
|
|
|
|
| 14 |
"boolq",
|
| 15 |
+
"commonsense_qa",
|
| 16 |
"copa",
|
| 17 |
+
"hellaswag",
|
| 18 |
"lambada_openai",
|
| 19 |
+
"openbookqa",
|
| 20 |
+
"piqa",
|
| 21 |
+
"rte",
|
| 22 |
+
"sciq",
|
| 23 |
+
"winogrande"
|
| 24 |
],
|
| 25 |
"tasks": {
|
| 26 |
"hellaswag": {
|
|
|
|
| 338 |
-0.08
|
| 339 |
],
|
| 340 |
"ci": NaN
|
| 341 |
+
},
|
| 342 |
+
"rte": {
|
| 343 |
+
"acc_norm": null,
|
| 344 |
+
"acc_norm_ci": 0.0,
|
| 345 |
+
"acc_norm_runs": [],
|
| 346 |
+
"acc": 0.5270758122743683,
|
| 347 |
+
"acc_ci": NaN,
|
| 348 |
+
"acc_runs": [
|
| 349 |
+
0.5270758122743683,
|
| 350 |
+
0.5270758122743683,
|
| 351 |
+
0.5270758122743683,
|
| 352 |
+
0.5270758122743683,
|
| 353 |
+
0.5270758122743683
|
| 354 |
+
],
|
| 355 |
+
"perplexity": null,
|
| 356 |
+
"perplexity_ci": 0.0,
|
| 357 |
+
"perplexity_runs": [],
|
| 358 |
+
"stderr": 0.030052303463143738,
|
| 359 |
+
"centered_acc": 0.054151624548736566,
|
| 360 |
+
"centered_acc_ci": NaN,
|
| 361 |
+
"centered_acc_runs": [
|
| 362 |
+
0.054151624548736566,
|
| 363 |
+
0.054151624548736566,
|
| 364 |
+
0.054151624548736566,
|
| 365 |
+
0.054151624548736566,
|
| 366 |
+
0.054151624548736566
|
| 367 |
+
],
|
| 368 |
+
"ci": NaN
|
| 369 |
+
},
|
| 370 |
+
"commonsense_qa": {
|
| 371 |
+
"acc_norm": null,
|
| 372 |
+
"acc_norm_ci": 0.0,
|
| 373 |
+
"acc_norm_runs": [],
|
| 374 |
+
"acc": 0.19574119574119575,
|
| 375 |
+
"acc_ci": NaN,
|
| 376 |
+
"acc_runs": [
|
| 377 |
+
0.19574119574119575,
|
| 378 |
+
0.19574119574119575,
|
| 379 |
+
0.19574119574119575,
|
| 380 |
+
0.19574119574119575,
|
| 381 |
+
0.19574119574119575
|
| 382 |
+
],
|
| 383 |
+
"perplexity": null,
|
| 384 |
+
"perplexity_ci": 0.0,
|
| 385 |
+
"perplexity_runs": [],
|
| 386 |
+
"stderr": 0.011359497363584397,
|
| 387 |
+
"centered_acc": null,
|
| 388 |
+
"centered_acc_ci": 0.0,
|
| 389 |
+
"centered_acc_runs": [],
|
| 390 |
+
"ci": NaN
|
| 391 |
}
|
| 392 |
},
|
| 393 |
"summary": {
|
| 394 |
+
"mean_acc_norm": 0.3023447895661438,
|
| 395 |
+
"mean_acc_norm_ci": 0.04070480970435736,
|
| 396 |
+
"mean_acc": 0.3255115382652869,
|
| 397 |
+
"mean_acc_ci": 0.04788953532303819,
|
| 398 |
+
"mean_centered_acc": 0.007725273053538176,
|
| 399 |
+
"mean_centered_acc_ci": 0.02910951315106165,
|
| 400 |
"mean_acc_norm_runs": [
|
| 401 |
+
0.2606054570802629,
|
| 402 |
+
0.2606054570802629,
|
| 403 |
+
0.2606054570802629,
|
| 404 |
+
0.2606054570802629,
|
| 405 |
+
0.2606054570802629,
|
| 406 |
+
0.530467899891186,
|
| 407 |
+
0.530467899891186,
|
| 408 |
+
0.530467899891186,
|
| 409 |
+
0.530467899891186,
|
| 410 |
+
0.530467899891186,
|
| 411 |
+
0.2588383838383838,
|
| 412 |
+
0.2588383838383838,
|
| 413 |
+
0.2588383838383838,
|
| 414 |
+
0.2588383838383838,
|
| 415 |
+
0.2588383838383838,
|
| 416 |
+
0.2781569965870307,
|
| 417 |
+
0.2781569965870307,
|
| 418 |
+
0.2781569965870307,
|
| 419 |
+
0.2781569965870307,
|
| 420 |
+
0.2781569965870307,
|
| 421 |
+
0.296,
|
| 422 |
+
0.296,
|
| 423 |
+
0.296,
|
| 424 |
+
0.296,
|
| 425 |
+
0.296,
|
| 426 |
+
0.19,
|
| 427 |
+
0.19,
|
| 428 |
+
0.19,
|
| 429 |
+
0.19,
|
| 430 |
+
0.19
|
| 431 |
],
|
| 432 |
"mean_acc_runs": [
|
| 433 |
+
0.25791674965146383,
|
| 434 |
+
0.25791674965146383,
|
| 435 |
+
0.25791674965146383,
|
| 436 |
+
0.25791674965146383,
|
| 437 |
+
0.25791674965146383,
|
| 438 |
+
0.5500544069640914,
|
| 439 |
+
0.5500544069640914,
|
| 440 |
+
0.5500544069640914,
|
| 441 |
+
0.5500544069640914,
|
| 442 |
+
0.5500544069640914,
|
| 443 |
+
0.2529461279461279,
|
| 444 |
+
0.2529461279461279,
|
| 445 |
+
0.2529461279461279,
|
| 446 |
+
0.2529461279461279,
|
| 447 |
+
0.2529461279461279,
|
| 448 |
+
0.22781569965870307,
|
| 449 |
+
0.22781569965870307,
|
| 450 |
+
0.22781569965870307,
|
| 451 |
+
0.22781569965870307,
|
| 452 |
+
0.22781569965870307,
|
| 453 |
+
0.166,
|
| 454 |
+
0.166,
|
| 455 |
+
0.166,
|
| 456 |
+
0.166,
|
| 457 |
+
0.166,
|
| 458 |
+
0.500394632991318,
|
| 459 |
+
0.500394632991318,
|
| 460 |
+
0.500394632991318,
|
| 461 |
+
0.500394632991318,
|
| 462 |
+
0.500394632991318,
|
| 463 |
+
0.3782874617737003,
|
| 464 |
+
0.3782874617737003,
|
| 465 |
+
0.3782874617737003,
|
| 466 |
+
0.3782874617737003,
|
| 467 |
+
0.3782874617737003,
|
| 468 |
+
0.58,
|
| 469 |
+
0.58,
|
| 470 |
+
0.58,
|
| 471 |
+
0.58,
|
| 472 |
+
0.58,
|
| 473 |
+
0.0,
|
| 474 |
+
0.0,
|
| 475 |
+
0.0,
|
| 476 |
+
0.0,
|
| 477 |
+
0.0,
|
| 478 |
+
0.053,
|
| 479 |
+
0.053,
|
| 480 |
+
0.053,
|
| 481 |
+
0.053,
|
| 482 |
+
0.053,
|
| 483 |
+
0.5270758122743683,
|
| 484 |
+
0.5270758122743683,
|
| 485 |
+
0.5270758122743683,
|
| 486 |
+
0.5270758122743683,
|
| 487 |
+
0.5270758122743683,
|
| 488 |
+
0.19574119574119575,
|
| 489 |
+
0.19574119574119575,
|
| 490 |
+
0.19574119574119575,
|
| 491 |
+
0.19574119574119575,
|
| 492 |
+
0.19574119574119575,
|
| 493 |
+
0.5424179104477612,
|
| 494 |
+
0.5424179104477612,
|
| 495 |
+
0.5424179104477612,
|
| 496 |
+
0.5424179104477612,
|
| 497 |
+
0.5424179104477612
|
| 498 |
],
|
| 499 |
"mean_centered_acc_runs": [
|
| 500 |
+
0.014140609440350524,
|
| 501 |
+
0.014140609440350524,
|
| 502 |
+
0.014140609440350524,
|
| 503 |
+
0.014140609440350524,
|
| 504 |
+
0.014140609440350524,
|
| 505 |
+
0.06093579978237207,
|
| 506 |
+
0.06093579978237207,
|
| 507 |
+
0.06093579978237207,
|
| 508 |
+
0.06093579978237207,
|
| 509 |
+
0.06093579978237207,
|
| 510 |
+
0.01178451178451175,
|
| 511 |
+
0.01178451178451175,
|
| 512 |
+
0.01178451178451175,
|
| 513 |
+
0.01178451178451175,
|
| 514 |
+
0.01178451178451175,
|
| 515 |
+
0.037542662116040924,
|
| 516 |
+
0.037542662116040924,
|
| 517 |
+
0.037542662116040924,
|
| 518 |
+
0.037542662116040924,
|
| 519 |
+
0.037542662116040924,
|
| 520 |
+
0.061333333333333316,
|
| 521 |
+
0.061333333333333316,
|
| 522 |
+
0.061333333333333316,
|
| 523 |
+
0.061333333333333316,
|
| 524 |
+
0.061333333333333316,
|
| 525 |
+
0.0007892659826360671,
|
| 526 |
+
0.0007892659826360671,
|
| 527 |
+
0.0007892659826360671,
|
| 528 |
+
0.0007892659826360671,
|
| 529 |
+
0.0007892659826360671,
|
| 530 |
+
-0.24342507645259936,
|
| 531 |
+
-0.24342507645259936,
|
| 532 |
+
-0.24342507645259936,
|
| 533 |
+
-0.24342507645259936,
|
| 534 |
+
-0.24342507645259936,
|
| 535 |
+
0.15999999999999992,
|
| 536 |
+
0.15999999999999992,
|
| 537 |
+
0.15999999999999992,
|
| 538 |
+
0.15999999999999992,
|
| 539 |
+
0.15999999999999992,
|
| 540 |
+
-0.08,
|
| 541 |
+
-0.08,
|
| 542 |
+
-0.08,
|
| 543 |
+
-0.08,
|
| 544 |
+
-0.08,
|
| 545 |
+
0.054151624548736566,
|
| 546 |
+
0.054151624548736566,
|
| 547 |
+
0.054151624548736566,
|
| 548 |
+
0.054151624548736566,
|
| 549 |
+
0.054151624548736566
|
| 550 |
],
|
| 551 |
+
"ci": 0.04070480970435736
|
| 552 |
},
|
| 553 |
"batch_size_per_device": 8,
|
| 554 |
"device": "cuda:0",
|
|
|
|
| 558 |
"tokenizer_has_l2r_pred": true,
|
| 559 |
"tokenizer_has_next_1_pred": true,
|
| 560 |
"max_next_i_train": 5
|
| 561 |
+
}
|
results/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024/zero_shot_eval/summary.jsonl
CHANGED
|
@@ -1 +1,3 @@
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024", "run": "random15-l2r50-i5-exp-fulle-lm-wsd-from-21024", "checkpoint_path": "/workspace/concept_model/runs/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024/checkpoint-21960", "checkpoint": "checkpoint-21960", "global_step": 21960, "tokens_seen": 23026728960, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2606054570802629, "acc_norm_ci": NaN, "acc_norm_runs": [0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629], "acc": 0.25791674965146383, "acc_ci": NaN, "acc_runs": [0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004380678585341082, "centered_acc": 0.014140609440350524, "centered_acc_ci": NaN, "centered_acc_runs": [0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524], "ci": NaN}, "piqa": {"acc_norm": 0.530467899891186, "acc_norm_ci": NaN, "acc_norm_runs": [0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186], "acc": 0.5500544069640914, "acc_ci": NaN, "acc_runs": [0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011644145418355097, "centered_acc": 0.06093579978237207, "centered_acc_ci": NaN, "centered_acc_runs": [0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207], "ci": NaN}, "arc_easy": {"acc_norm": 0.2588383838383838, "acc_norm_ci": NaN, "acc_norm_runs": [0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838], "acc": 0.2529461279461279, "acc_ci": NaN, "acc_runs": [0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008987501845757919, "centered_acc": 0.01178451178451175, "centered_acc_ci": NaN, "centered_acc_runs": [0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175], "ci": NaN}, "arc_challenge": {"acc_norm": 0.2781569965870307, "acc_norm_ci": NaN, "acc_norm_runs": [0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307], "acc": 0.22781569965870307, "acc_ci": NaN, "acc_runs": [0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.013094469919538897, "centered_acc": 0.037542662116040924, "centered_acc_ci": NaN, "centered_acc_runs": [0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924], "ci": NaN}, "openbookqa": {"acc_norm": 0.296, "acc_norm_ci": NaN, "acc_norm_runs": [0.296, 0.296, 0.296, 0.296, 0.296], "acc": 0.166, "acc_ci": NaN, "acc_runs": [0.166, 0.166, 0.166, 0.166, 0.166], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020435342091896222, "centered_acc": 0.061333333333333316, "centered_acc_ci": NaN, "centered_acc_runs": [0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.500394632991318, "acc_ci": NaN, "acc_runs": [0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052481306049611, "centered_acc": 0.0007892659826360671, "centered_acc_ci": NaN, "centered_acc_runs": [0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.58, "acc_ci": NaN, "acc_runs": [0.58, 0.58, 0.58, 0.58, 0.58], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.04960449637488582, "centered_acc": 0.15999999999999992, "centered_acc_ci": NaN, "centered_acc_runs": [0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 9823740.063432544, "perplexity_ci": NaN, "perplexity_runs": [9823740.063432544, 9823740.063432544, 9823740.063432544, 9823740.063432544, 9823740.063432544], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.19, "acc_norm_ci": NaN, "acc_norm_runs": [0.19, 0.19, 0.19, 0.19, 0.19], "acc": 0.053000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.053, 0.053, 0.053, 0.053, 0.053], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012411851354816256, "centered_acc": -0.08, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08, -0.08, -0.08, -0.08, -0.08], "ci": NaN}}, "summary": {"mean_acc_norm": 0.3023447895661439, "mean_acc_norm_ci": NaN, "mean_acc": 0.29664150789854044, "mean_acc_ci": NaN, "mean_centered_acc": 0.0025667895540716894, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.3023447895661439, 0.3023447895661439, 0.3023447895661439, 0.3023447895661439, 0.3023447895661439], "mean_acc_runs": [0.29664150789854044, 0.29664150789854044, 0.29664150789854044, 0.29664150789854044, 0.29664150789854044], "mean_centered_acc_runs": [0.0025667895540716894, 0.0025667895540716894, 0.0025667895540716894, 0.0025667895540716894, 0.0025667895540716894], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
|
|
|
|
|
|
|
|
| 1 |
{"run_dir": "/workspace/concept_model/runs/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024", "run": "random15-l2r50-i5-exp-fulle-lm-wsd-from-21024", "checkpoint_path": "/workspace/concept_model/runs/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024/checkpoint-21960", "checkpoint": "checkpoint-21960", "global_step": 21960, "tokens_seen": 23026728960, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["hellaswag", "piqa", "arc_easy", "arc_challenge", "openbookqa", "winogrande", "boolq", "copa", "lambada_openai", "sciq"], "tasks": {"hellaswag": {"acc_norm": 0.2606054570802629, "acc_norm_ci": NaN, "acc_norm_runs": [0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629], "acc": 0.25791674965146383, "acc_ci": NaN, "acc_runs": [0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004380678585341082, "centered_acc": 0.014140609440350524, "centered_acc_ci": NaN, "centered_acc_runs": [0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524], "ci": NaN}, "piqa": {"acc_norm": 0.530467899891186, "acc_norm_ci": NaN, "acc_norm_runs": [0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186], "acc": 0.5500544069640914, "acc_ci": NaN, "acc_runs": [0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011644145418355097, "centered_acc": 0.06093579978237207, "centered_acc_ci": NaN, "centered_acc_runs": [0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207], "ci": NaN}, "arc_easy": {"acc_norm": 0.2588383838383838, "acc_norm_ci": NaN, "acc_norm_runs": [0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838], "acc": 0.2529461279461279, "acc_ci": NaN, "acc_runs": [0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008987501845757919, "centered_acc": 0.01178451178451175, "centered_acc_ci": NaN, "centered_acc_runs": [0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175], "ci": NaN}, "arc_challenge": {"acc_norm": 0.2781569965870307, "acc_norm_ci": NaN, "acc_norm_runs": [0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307], "acc": 0.22781569965870307, "acc_ci": NaN, "acc_runs": [0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.013094469919538897, "centered_acc": 0.037542662116040924, "centered_acc_ci": NaN, "centered_acc_runs": [0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924], "ci": NaN}, "openbookqa": {"acc_norm": 0.296, "acc_norm_ci": NaN, "acc_norm_runs": [0.296, 0.296, 0.296, 0.296, 0.296], "acc": 0.166, "acc_ci": NaN, "acc_runs": [0.166, 0.166, 0.166, 0.166, 0.166], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020435342091896222, "centered_acc": 0.061333333333333316, "centered_acc_ci": NaN, "centered_acc_runs": [0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.500394632991318, "acc_ci": NaN, "acc_runs": [0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052481306049611, "centered_acc": 0.0007892659826360671, "centered_acc_ci": NaN, "centered_acc_runs": [0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.58, "acc_ci": NaN, "acc_runs": [0.58, 0.58, 0.58, 0.58, 0.58], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.04960449637488582, "centered_acc": 0.15999999999999992, "centered_acc_ci": NaN, "centered_acc_runs": [0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 9823740.063432544, "perplexity_ci": NaN, "perplexity_runs": [9823740.063432544, 9823740.063432544, 9823740.063432544, 9823740.063432544, 9823740.063432544], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.19, "acc_norm_ci": NaN, "acc_norm_runs": [0.19, 0.19, 0.19, 0.19, 0.19], "acc": 0.053000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.053, 0.053, 0.053, 0.053, 0.053], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012411851354816256, "centered_acc": -0.08, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08, -0.08, -0.08, -0.08, -0.08], "ci": NaN}}, "summary": {"mean_acc_norm": 0.3023447895661439, "mean_acc_norm_ci": NaN, "mean_acc": 0.29664150789854044, "mean_acc_ci": NaN, "mean_centered_acc": 0.0025667895540716894, "mean_centered_acc_ci": NaN, "mean_acc_norm_runs": [0.3023447895661439, 0.3023447895661439, 0.3023447895661439, 0.3023447895661439, 0.3023447895661439], "mean_acc_runs": [0.29664150789854044, 0.29664150789854044, 0.29664150789854044, 0.29664150789854044, 0.29664150789854044], "mean_centered_acc_runs": [0.0025667895540716894, 0.0025667895540716894, 0.0025667895540716894, 0.0025667895540716894, 0.0025667895540716894], "ci": NaN}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
| 2 |
+
{"run_dir": "/workspace/concept_model/runs/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024", "run": "random15-l2r50-i5-exp-fulle-lm-wsd-from-21024", "checkpoint_path": "/workspace/concept_model/runs/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024/checkpoint-21960", "checkpoint": "checkpoint-21960", "global_step": 21960, "tokens_seen": 23026728960, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2606054570802629, "acc_norm_ci": NaN, "acc_norm_runs": [0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629], "acc": 0.25791674965146383, "acc_ci": NaN, "acc_runs": [0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004380678585341082, "centered_acc": 0.014140609440350524, "centered_acc_ci": NaN, "centered_acc_runs": [0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524], "ci": NaN}, "piqa": {"acc_norm": 0.530467899891186, "acc_norm_ci": NaN, "acc_norm_runs": [0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186], "acc": 0.5500544069640914, "acc_ci": NaN, "acc_runs": [0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011644145418355097, "centered_acc": 0.06093579978237207, "centered_acc_ci": NaN, "centered_acc_runs": [0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207], "ci": NaN}, "arc_easy": {"acc_norm": 0.2588383838383838, "acc_norm_ci": NaN, "acc_norm_runs": [0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838], "acc": 0.2529461279461279, "acc_ci": NaN, "acc_runs": [0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008987501845757919, "centered_acc": 0.01178451178451175, "centered_acc_ci": NaN, "centered_acc_runs": [0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175], "ci": NaN}, "arc_challenge": {"acc_norm": 0.2781569965870307, "acc_norm_ci": NaN, "acc_norm_runs": [0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307], "acc": 0.22781569965870307, "acc_ci": NaN, "acc_runs": [0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.013094469919538897, "centered_acc": 0.037542662116040924, "centered_acc_ci": NaN, "centered_acc_runs": [0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924], "ci": NaN}, "openbookqa": {"acc_norm": 0.296, "acc_norm_ci": NaN, "acc_norm_runs": [0.296, 0.296, 0.296, 0.296, 0.296], "acc": 0.166, "acc_ci": NaN, "acc_runs": [0.166, 0.166, 0.166, 0.166, 0.166], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020435342091896222, "centered_acc": 0.061333333333333316, "centered_acc_ci": NaN, "centered_acc_runs": [0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.500394632991318, "acc_ci": NaN, "acc_runs": [0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052481306049611, "centered_acc": 0.0007892659826360671, "centered_acc_ci": NaN, "centered_acc_runs": [0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.58, "acc_ci": NaN, "acc_runs": [0.58, 0.58, 0.58, 0.58, 0.58], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.04960449637488582, "centered_acc": 0.15999999999999992, "centered_acc_ci": NaN, "centered_acc_runs": [0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 9823740.063432544, "perplexity_ci": NaN, "perplexity_runs": [9823740.063432544, 9823740.063432544, 9823740.063432544, 9823740.063432544, 9823740.063432544], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.19, "acc_norm_ci": NaN, "acc_norm_runs": [0.19, 0.19, 0.19, 0.19, 0.19], "acc": 0.053000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.053, 0.053, 0.053, 0.053, 0.053], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012411851354816256, "centered_acc": -0.08, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08, -0.08, -0.08, -0.08, -0.08], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.3023447895661438, "mean_acc_norm_ci": 0.04070480970435736, "mean_acc": 0.3074360072500807, "mean_acc_ci": 0.049149250995015215, "mean_centered_acc": 0.007725273053538176, "mean_centered_acc_ci": 0.02910951315106165, "mean_acc_norm_runs": [0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.296, 0.296, 0.296, 0.296, 0.296, 0.19, 0.19, 0.19, 0.19, 0.19], "mean_acc_runs": [0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.166, 0.166, 0.166, 0.166, 0.166, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.58, 0.58, 0.58, 0.58, 0.58, 0.0, 0.0, 0.0, 0.0, 0.0, 0.053, 0.053, 0.053, 0.053, 0.053, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "mean_centered_acc_runs": [0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992, -0.08, -0.08, -0.08, -0.08, -0.08, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.04070480970435736}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
| 3 |
+
{"run_dir": "/workspace/concept_model/runs/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024", "run": "random15-l2r50-i5-exp-fulle-lm-wsd-from-21024", "checkpoint_path": "/workspace/concept_model/runs/random15-l2r50-i5-exp-fulle-lm-wsd-from-21024/checkpoint-21960", "checkpoint": "checkpoint-21960", "global_step": 21960, "tokens_seen": 23026728960, "num_fewshot": 0, "runs_per_task": 5, "base_seed": 42, "tasks_evaluated": ["arc_challenge", "arc_easy", "blimp", "boolq", "commonsense_qa", "copa", "hellaswag", "lambada_openai", "openbookqa", "piqa", "rte", "sciq", "winogrande"], "tasks": {"hellaswag": {"acc_norm": 0.2606054570802629, "acc_norm_ci": NaN, "acc_norm_runs": [0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629], "acc": 0.25791674965146383, "acc_ci": NaN, "acc_runs": [0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.004380678585341082, "centered_acc": 0.014140609440350524, "centered_acc_ci": NaN, "centered_acc_runs": [0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524], "ci": NaN}, "piqa": {"acc_norm": 0.530467899891186, "acc_norm_ci": NaN, "acc_norm_runs": [0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186], "acc": 0.5500544069640914, "acc_ci": NaN, "acc_runs": [0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011644145418355097, "centered_acc": 0.06093579978237207, "centered_acc_ci": NaN, "centered_acc_runs": [0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207], "ci": NaN}, "arc_easy": {"acc_norm": 0.2588383838383838, "acc_norm_ci": NaN, "acc_norm_runs": [0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838], "acc": 0.2529461279461279, "acc_ci": NaN, "acc_runs": [0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008987501845757919, "centered_acc": 0.01178451178451175, "centered_acc_ci": NaN, "centered_acc_runs": [0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175], "ci": NaN}, "arc_challenge": {"acc_norm": 0.2781569965870307, "acc_norm_ci": NaN, "acc_norm_runs": [0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307], "acc": 0.22781569965870307, "acc_ci": NaN, "acc_runs": [0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.013094469919538897, "centered_acc": 0.037542662116040924, "centered_acc_ci": NaN, "centered_acc_runs": [0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924], "ci": NaN}, "openbookqa": {"acc_norm": 0.296, "acc_norm_ci": NaN, "acc_norm_runs": [0.296, 0.296, 0.296, 0.296, 0.296], "acc": 0.166, "acc_ci": NaN, "acc_runs": [0.166, 0.166, 0.166, 0.166, 0.166], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.020435342091896222, "centered_acc": 0.061333333333333316, "centered_acc_ci": NaN, "centered_acc_runs": [0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316], "ci": NaN}, "winogrande": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.500394632991318, "acc_ci": NaN, "acc_runs": [0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.014052481306049611, "centered_acc": 0.0007892659826360671, "centered_acc_ci": NaN, "centered_acc_runs": [0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671], "ci": NaN}, "boolq": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.3782874617737003, "acc_ci": NaN, "acc_runs": [0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.008482001133931097, "centered_acc": -0.24342507645259936, "centered_acc_ci": NaN, "centered_acc_runs": [-0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936], "ci": NaN}, "copa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.58, "acc_ci": NaN, "acc_runs": [0.58, 0.58, 0.58, 0.58, 0.58], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.04960449637488582, "centered_acc": 0.15999999999999992, "centered_acc_ci": NaN, "centered_acc_runs": [0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992], "ci": NaN}, "lambada_openai": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.0, "acc_ci": NaN, "acc_runs": [0.0, 0.0, 0.0, 0.0, 0.0], "perplexity": 9823740.063432544, "perplexity_ci": NaN, "perplexity_runs": [9823740.063432544, 9823740.063432544, 9823740.063432544, 9823740.063432544, 9823740.063432544], "stderr": 0.0, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "sciq": {"acc_norm": 0.19, "acc_norm_ci": NaN, "acc_norm_runs": [0.19, 0.19, 0.19, 0.19, 0.19], "acc": 0.053000000000000005, "acc_ci": 6.938893903907228e-18, "acc_runs": [0.053, 0.053, 0.053, 0.053, 0.053], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.012411851354816256, "centered_acc": -0.08, "centered_acc_ci": NaN, "centered_acc_runs": [-0.08, -0.08, -0.08, -0.08, -0.08], "ci": NaN}, "rte": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5270758122743683, "acc_ci": NaN, "acc_runs": [0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.030052303463143738, "centered_acc": 0.054151624548736566, "centered_acc_ci": NaN, "centered_acc_runs": [0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": NaN}, "commonsense_qa": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.19574119574119575, "acc_ci": NaN, "acc_runs": [0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.011359497363584397, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}, "blimp": {"acc_norm": null, "acc_norm_ci": 0.0, "acc_norm_runs": [], "acc": 0.5424179104477612, "acc_ci": NaN, "acc_runs": [0.5424179104477612, 0.5424179104477612, 0.5424179104477612, 0.5424179104477612, 0.5424179104477612], "perplexity": null, "perplexity_ci": 0.0, "perplexity_runs": [], "stderr": 0.001841859059478856, "centered_acc": null, "centered_acc_ci": 0.0, "centered_acc_runs": [], "ci": NaN}}, "summary": {"mean_acc_norm": 0.3023447895661438, "mean_acc_norm_ci": 0.04070480970435736, "mean_acc": 0.3255115382652869, "mean_acc_ci": 0.04788953532303819, "mean_centered_acc": 0.007725273053538176, "mean_centered_acc_ci": 0.02910951315106165, "mean_acc_norm_runs": [0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.2606054570802629, 0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186, 0.530467899891186, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2588383838383838, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.2781569965870307, 0.296, 0.296, 0.296, 0.296, 0.296, 0.19, 0.19, 0.19, 0.19, 0.19], "mean_acc_runs": [0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.25791674965146383, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.5500544069640914, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.2529461279461279, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.22781569965870307, 0.166, 0.166, 0.166, 0.166, 0.166, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.500394632991318, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.3782874617737003, 0.58, 0.58, 0.58, 0.58, 0.58, 0.0, 0.0, 0.0, 0.0, 0.0, 0.053, 0.053, 0.053, 0.053, 0.053, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.5270758122743683, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.19574119574119575, 0.5424179104477612, 0.5424179104477612, 0.5424179104477612, 0.5424179104477612, 0.5424179104477612], "mean_centered_acc_runs": [0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.014140609440350524, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.06093579978237207, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.01178451178451175, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.037542662116040924, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.061333333333333316, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, 0.0007892659826360671, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, -0.24342507645259936, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992, 0.15999999999999992, -0.08, -0.08, -0.08, -0.08, -0.08, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566, 0.054151624548736566], "ci": 0.04070480970435736}, "batch_size_per_device": 8, "device": "cuda:0", "dtype": "float16", "add_l2r_token": true, "add_next_1_pred_token": true, "tokenizer_has_l2r_pred": true, "tokenizer_has_next_1_pred": true, "max_next_i_train": 5}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-1152.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-1152",
|
| 4 |
+
"checkpoint": "checkpoint-1152",
|
| 5 |
+
"global_step": 1152,
|
| 6 |
+
"eval_loss": 3.9638848304748535,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.9638848304748535,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.0317,
|
| 11 |
+
"eval_samples_per_second": 16.722,
|
| 12 |
+
"eval_steps_per_second": 8.361,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-1440.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-1440",
|
| 4 |
+
"checkpoint": "checkpoint-1440",
|
| 5 |
+
"global_step": 1440,
|
| 6 |
+
"eval_loss": 3.9123761653900146,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.9123761653900146,
|
| 9 |
+
"eval_model_preparation_time": 0.0011,
|
| 10 |
+
"eval_runtime": 146.0584,
|
| 11 |
+
"eval_samples_per_second": 16.719,
|
| 12 |
+
"eval_steps_per_second": 8.36,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-1728.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-1728",
|
| 4 |
+
"checkpoint": "checkpoint-1728",
|
| 5 |
+
"global_step": 1728,
|
| 6 |
+
"eval_loss": 3.915787696838379,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.915787696838379,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.1166,
|
| 11 |
+
"eval_samples_per_second": 16.713,
|
| 12 |
+
"eval_steps_per_second": 8.356,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-2016.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-2016",
|
| 4 |
+
"checkpoint": "checkpoint-2016",
|
| 5 |
+
"global_step": 2016,
|
| 6 |
+
"eval_loss": 3.9413511753082275,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.9413511753082275,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.1224,
|
| 11 |
+
"eval_samples_per_second": 16.712,
|
| 12 |
+
"eval_steps_per_second": 8.356,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-2304.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-2304",
|
| 4 |
+
"checkpoint": "checkpoint-2304",
|
| 5 |
+
"global_step": 2304,
|
| 6 |
+
"eval_loss": 3.978717565536499,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.978717565536499,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.1275,
|
| 11 |
+
"eval_samples_per_second": 16.711,
|
| 12 |
+
"eval_steps_per_second": 8.356,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-2592.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-2592",
|
| 4 |
+
"checkpoint": "checkpoint-2592",
|
| 5 |
+
"global_step": 2592,
|
| 6 |
+
"eval_loss": 4.025238037109375,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.025238037109375,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.2964,
|
| 11 |
+
"eval_samples_per_second": 16.692,
|
| 12 |
+
"eval_steps_per_second": 8.346,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-288.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-288",
|
| 4 |
+
"checkpoint": "checkpoint-288",
|
| 5 |
+
"global_step": 288,
|
| 6 |
+
"eval_loss": 5.638572692871094,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 5.638572692871094,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 145.6811,
|
| 11 |
+
"eval_samples_per_second": 16.763,
|
| 12 |
+
"eval_steps_per_second": 8.381,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-2880.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-2880",
|
| 4 |
+
"checkpoint": "checkpoint-2880",
|
| 5 |
+
"global_step": 2880,
|
| 6 |
+
"eval_loss": 4.071658611297607,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.071658611297607,
|
| 9 |
+
"eval_model_preparation_time": 0.0011,
|
| 10 |
+
"eval_runtime": 146.3406,
|
| 11 |
+
"eval_samples_per_second": 16.687,
|
| 12 |
+
"eval_steps_per_second": 8.344,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-3168.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-3168",
|
| 4 |
+
"checkpoint": "checkpoint-3168",
|
| 5 |
+
"global_step": 3168,
|
| 6 |
+
"eval_loss": 4.124143600463867,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.124143600463867,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.3235,
|
| 11 |
+
"eval_samples_per_second": 16.689,
|
| 12 |
+
"eval_steps_per_second": 8.345,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-3456.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-3456",
|
| 4 |
+
"checkpoint": "checkpoint-3456",
|
| 5 |
+
"global_step": 3456,
|
| 6 |
+
"eval_loss": 4.178173065185547,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.178173065185547,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.1676,
|
| 11 |
+
"eval_samples_per_second": 16.707,
|
| 12 |
+
"eval_steps_per_second": 8.353,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-3744.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-3744",
|
| 4 |
+
"checkpoint": "checkpoint-3744",
|
| 5 |
+
"global_step": 3744,
|
| 6 |
+
"eval_loss": 4.225054740905762,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.225054740905762,
|
| 9 |
+
"eval_model_preparation_time": 0.0013,
|
| 10 |
+
"eval_runtime": 148.8213,
|
| 11 |
+
"eval_samples_per_second": 16.409,
|
| 12 |
+
"eval_steps_per_second": 8.204,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-4032.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-4032",
|
| 4 |
+
"checkpoint": "checkpoint-4032",
|
| 5 |
+
"global_step": 4032,
|
| 6 |
+
"eval_loss": 4.26896333694458,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.26896333694458,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.5863,
|
| 11 |
+
"eval_samples_per_second": 16.546,
|
| 12 |
+
"eval_steps_per_second": 8.273,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-4320.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-4320",
|
| 4 |
+
"checkpoint": "checkpoint-4320",
|
| 5 |
+
"global_step": 4320,
|
| 6 |
+
"eval_loss": 4.324903964996338,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.324903964996338,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.5973,
|
| 11 |
+
"eval_samples_per_second": 16.545,
|
| 12 |
+
"eval_steps_per_second": 8.273,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-4608.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-4608",
|
| 4 |
+
"checkpoint": "checkpoint-4608",
|
| 5 |
+
"global_step": 4608,
|
| 6 |
+
"eval_loss": 4.351988315582275,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.351988315582275,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.6417,
|
| 11 |
+
"eval_samples_per_second": 16.54,
|
| 12 |
+
"eval_steps_per_second": 8.27,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-4896.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-4896",
|
| 4 |
+
"checkpoint": "checkpoint-4896",
|
| 5 |
+
"global_step": 4896,
|
| 6 |
+
"eval_loss": 4.411014080047607,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.411014080047607,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.6517,
|
| 11 |
+
"eval_samples_per_second": 16.539,
|
| 12 |
+
"eval_steps_per_second": 8.269,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-5184.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-5184",
|
| 4 |
+
"checkpoint": "checkpoint-5184",
|
| 5 |
+
"global_step": 5184,
|
| 6 |
+
"eval_loss": 4.438808917999268,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.438808917999268,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.6542,
|
| 11 |
+
"eval_samples_per_second": 16.539,
|
| 12 |
+
"eval_steps_per_second": 8.269,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-5472.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-5472",
|
| 4 |
+
"checkpoint": "checkpoint-5472",
|
| 5 |
+
"global_step": 5472,
|
| 6 |
+
"eval_loss": 4.491748332977295,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.491748332977295,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.616,
|
| 11 |
+
"eval_samples_per_second": 16.543,
|
| 12 |
+
"eval_steps_per_second": 8.271,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-576.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-576",
|
| 4 |
+
"checkpoint": "checkpoint-576",
|
| 5 |
+
"global_step": 576,
|
| 6 |
+
"eval_loss": 4.557515621185303,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.557515621185303,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 145.8496,
|
| 11 |
+
"eval_samples_per_second": 16.743,
|
| 12 |
+
"eval_steps_per_second": 8.372,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-5760.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-5760",
|
| 4 |
+
"checkpoint": "checkpoint-5760",
|
| 5 |
+
"global_step": 5760,
|
| 6 |
+
"eval_loss": 4.511440277099609,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.511440277099609,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.6209,
|
| 11 |
+
"eval_samples_per_second": 16.542,
|
| 12 |
+
"eval_steps_per_second": 8.271,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-6048.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-6048",
|
| 4 |
+
"checkpoint": "checkpoint-6048",
|
| 5 |
+
"global_step": 6048,
|
| 6 |
+
"eval_loss": 4.539799690246582,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.539799690246582,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.6282,
|
| 11 |
+
"eval_samples_per_second": 16.542,
|
| 12 |
+
"eval_steps_per_second": 8.271,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-6336.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-6336",
|
| 4 |
+
"checkpoint": "checkpoint-6336",
|
| 5 |
+
"global_step": 6336,
|
| 6 |
+
"eval_loss": 4.580272674560547,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.580272674560547,
|
| 9 |
+
"eval_model_preparation_time": 0.0013,
|
| 10 |
+
"eval_runtime": 147.635,
|
| 11 |
+
"eval_samples_per_second": 16.541,
|
| 12 |
+
"eval_steps_per_second": 8.27,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-6624.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-6624",
|
| 4 |
+
"checkpoint": "checkpoint-6624",
|
| 5 |
+
"global_step": 6624,
|
| 6 |
+
"eval_loss": 4.607502460479736,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.607502460479736,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.6497,
|
| 11 |
+
"eval_samples_per_second": 16.539,
|
| 12 |
+
"eval_steps_per_second": 8.27,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-6912.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-6912",
|
| 4 |
+
"checkpoint": "checkpoint-6912",
|
| 5 |
+
"global_step": 6912,
|
| 6 |
+
"eval_loss": 4.647249698638916,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.647249698638916,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.6625,
|
| 11 |
+
"eval_samples_per_second": 16.538,
|
| 12 |
+
"eval_steps_per_second": 8.269,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-72.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-72",
|
| 4 |
+
"checkpoint": "checkpoint-72",
|
| 5 |
+
"global_step": 72,
|
| 6 |
+
"eval_loss": 7.4754109382629395,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 7.4754109382629395,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.8928,
|
| 11 |
+
"eval_samples_per_second": 16.624,
|
| 12 |
+
"eval_steps_per_second": 8.312,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-7200.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-7200",
|
| 4 |
+
"checkpoint": "checkpoint-7200",
|
| 5 |
+
"global_step": 7200,
|
| 6 |
+
"eval_loss": 4.675676345825195,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.675676345825195,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.6519,
|
| 11 |
+
"eval_samples_per_second": 16.539,
|
| 12 |
+
"eval_steps_per_second": 8.269,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/checkpoint-864.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-fulle-lm",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-864",
|
| 4 |
+
"checkpoint": "checkpoint-864",
|
| 5 |
+
"global_step": 864,
|
| 6 |
+
"eval_loss": 4.11744499206543,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.11744499206543,
|
| 9 |
+
"eval_model_preparation_time": 0.0013,
|
| 10 |
+
"eval_runtime": 146.056,
|
| 11 |
+
"eval_samples_per_second": 16.72,
|
| 12 |
+
"eval_steps_per_second": 8.36,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": false,
|
| 34 |
+
"max_next_i_train": 1
|
| 35 |
+
}
|
results/random5-fulle-lm/l2r_n1_eval/summary.jsonl
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-72", "checkpoint": "checkpoint-72", "global_step": 72, "eval_loss": 7.4754109382629395, "metrics": {"eval_loss": 7.4754109382629395, "eval_model_preparation_time": 0.0012, "eval_runtime": 146.8928, "eval_samples_per_second": 16.624, "eval_steps_per_second": 8.312, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 2 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-288", "checkpoint": "checkpoint-288", "global_step": 288, "eval_loss": 5.638572692871094, "metrics": {"eval_loss": 5.638572692871094, "eval_model_preparation_time": 0.0012, "eval_runtime": 145.6811, "eval_samples_per_second": 16.763, "eval_steps_per_second": 8.381, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 3 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-576", "checkpoint": "checkpoint-576", "global_step": 576, "eval_loss": 4.557515621185303, "metrics": {"eval_loss": 4.557515621185303, "eval_model_preparation_time": 0.0012, "eval_runtime": 145.8496, "eval_samples_per_second": 16.743, "eval_steps_per_second": 8.372, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 4 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-864", "checkpoint": "checkpoint-864", "global_step": 864, "eval_loss": 4.11744499206543, "metrics": {"eval_loss": 4.11744499206543, "eval_model_preparation_time": 0.0013, "eval_runtime": 146.056, "eval_samples_per_second": 16.72, "eval_steps_per_second": 8.36, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 5 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-1152", "checkpoint": "checkpoint-1152", "global_step": 1152, "eval_loss": 3.9638848304748535, "metrics": {"eval_loss": 3.9638848304748535, "eval_model_preparation_time": 0.0012, "eval_runtime": 146.0317, "eval_samples_per_second": 16.722, "eval_steps_per_second": 8.361, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 6 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-1440", "checkpoint": "checkpoint-1440", "global_step": 1440, "eval_loss": 3.9123761653900146, "metrics": {"eval_loss": 3.9123761653900146, "eval_model_preparation_time": 0.0011, "eval_runtime": 146.0584, "eval_samples_per_second": 16.719, "eval_steps_per_second": 8.36, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 7 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-1728", "checkpoint": "checkpoint-1728", "global_step": 1728, "eval_loss": 3.915787696838379, "metrics": {"eval_loss": 3.915787696838379, "eval_model_preparation_time": 0.0012, "eval_runtime": 146.1166, "eval_samples_per_second": 16.713, "eval_steps_per_second": 8.356, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 8 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-2016", "checkpoint": "checkpoint-2016", "global_step": 2016, "eval_loss": 3.9413511753082275, "metrics": {"eval_loss": 3.9413511753082275, "eval_model_preparation_time": 0.0012, "eval_runtime": 146.1224, "eval_samples_per_second": 16.712, "eval_steps_per_second": 8.356, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 9 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-2304", "checkpoint": "checkpoint-2304", "global_step": 2304, "eval_loss": 3.978717565536499, "metrics": {"eval_loss": 3.978717565536499, "eval_model_preparation_time": 0.0012, "eval_runtime": 146.1275, "eval_samples_per_second": 16.711, "eval_steps_per_second": 8.356, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 10 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-2592", "checkpoint": "checkpoint-2592", "global_step": 2592, "eval_loss": 4.025238037109375, "metrics": {"eval_loss": 4.025238037109375, "eval_model_preparation_time": 0.0012, "eval_runtime": 146.2964, "eval_samples_per_second": 16.692, "eval_steps_per_second": 8.346, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 11 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-2880", "checkpoint": "checkpoint-2880", "global_step": 2880, "eval_loss": 4.071658611297607, "metrics": {"eval_loss": 4.071658611297607, "eval_model_preparation_time": 0.0011, "eval_runtime": 146.3406, "eval_samples_per_second": 16.687, "eval_steps_per_second": 8.344, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 12 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-3168", "checkpoint": "checkpoint-3168", "global_step": 3168, "eval_loss": 4.124143600463867, "metrics": {"eval_loss": 4.124143600463867, "eval_model_preparation_time": 0.0012, "eval_runtime": 146.3235, "eval_samples_per_second": 16.689, "eval_steps_per_second": 8.345, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 13 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-3456", "checkpoint": "checkpoint-3456", "global_step": 3456, "eval_loss": 4.178173065185547, "metrics": {"eval_loss": 4.178173065185547, "eval_model_preparation_time": 0.0012, "eval_runtime": 146.1676, "eval_samples_per_second": 16.707, "eval_steps_per_second": 8.353, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 14 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-3744", "checkpoint": "checkpoint-3744", "global_step": 3744, "eval_loss": 4.225054740905762, "metrics": {"eval_loss": 4.225054740905762, "eval_model_preparation_time": 0.0013, "eval_runtime": 148.8213, "eval_samples_per_second": 16.409, "eval_steps_per_second": 8.204, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 15 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-4032", "checkpoint": "checkpoint-4032", "global_step": 4032, "eval_loss": 4.26896333694458, "metrics": {"eval_loss": 4.26896333694458, "eval_model_preparation_time": 0.0012, "eval_runtime": 147.5863, "eval_samples_per_second": 16.546, "eval_steps_per_second": 8.273, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 16 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-4320", "checkpoint": "checkpoint-4320", "global_step": 4320, "eval_loss": 4.324903964996338, "metrics": {"eval_loss": 4.324903964996338, "eval_model_preparation_time": 0.0012, "eval_runtime": 147.5973, "eval_samples_per_second": 16.545, "eval_steps_per_second": 8.273, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 17 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-4608", "checkpoint": "checkpoint-4608", "global_step": 4608, "eval_loss": 4.351988315582275, "metrics": {"eval_loss": 4.351988315582275, "eval_model_preparation_time": 0.0012, "eval_runtime": 147.6417, "eval_samples_per_second": 16.54, "eval_steps_per_second": 8.27, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 18 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-4896", "checkpoint": "checkpoint-4896", "global_step": 4896, "eval_loss": 4.411014080047607, "metrics": {"eval_loss": 4.411014080047607, "eval_model_preparation_time": 0.0012, "eval_runtime": 147.6517, "eval_samples_per_second": 16.539, "eval_steps_per_second": 8.269, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 19 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-5184", "checkpoint": "checkpoint-5184", "global_step": 5184, "eval_loss": 4.438808917999268, "metrics": {"eval_loss": 4.438808917999268, "eval_model_preparation_time": 0.0012, "eval_runtime": 147.6542, "eval_samples_per_second": 16.539, "eval_steps_per_second": 8.269, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 20 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-5472", "checkpoint": "checkpoint-5472", "global_step": 5472, "eval_loss": 4.491748332977295, "metrics": {"eval_loss": 4.491748332977295, "eval_model_preparation_time": 0.0012, "eval_runtime": 147.616, "eval_samples_per_second": 16.543, "eval_steps_per_second": 8.271, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 21 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-5760", "checkpoint": "checkpoint-5760", "global_step": 5760, "eval_loss": 4.511440277099609, "metrics": {"eval_loss": 4.511440277099609, "eval_model_preparation_time": 0.0012, "eval_runtime": 147.6209, "eval_samples_per_second": 16.542, "eval_steps_per_second": 8.271, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 22 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-6048", "checkpoint": "checkpoint-6048", "global_step": 6048, "eval_loss": 4.539799690246582, "metrics": {"eval_loss": 4.539799690246582, "eval_model_preparation_time": 0.0012, "eval_runtime": 147.6282, "eval_samples_per_second": 16.542, "eval_steps_per_second": 8.271, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 23 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-6336", "checkpoint": "checkpoint-6336", "global_step": 6336, "eval_loss": 4.580272674560547, "metrics": {"eval_loss": 4.580272674560547, "eval_model_preparation_time": 0.0013, "eval_runtime": 147.635, "eval_samples_per_second": 16.541, "eval_steps_per_second": 8.27, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 24 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-6624", "checkpoint": "checkpoint-6624", "global_step": 6624, "eval_loss": 4.607502460479736, "metrics": {"eval_loss": 4.607502460479736, "eval_model_preparation_time": 0.0012, "eval_runtime": 147.6497, "eval_samples_per_second": 16.539, "eval_steps_per_second": 8.27, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 25 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-6912", "checkpoint": "checkpoint-6912", "global_step": 6912, "eval_loss": 4.647249698638916, "metrics": {"eval_loss": 4.647249698638916, "eval_model_preparation_time": 0.0012, "eval_runtime": 147.6625, "eval_samples_per_second": 16.538, "eval_steps_per_second": 8.269, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 26 |
+
{"run_dir": "/workspace/concept_model/runs/random5-fulle-lm", "checkpoint_path": "/workspace/concept_model/runs/random5-fulle-lm/checkpoint-7200", "checkpoint": "checkpoint-7200", "global_step": 7200, "eval_loss": 4.675676345825195, "metrics": {"eval_loss": 4.675676345825195, "eval_model_preparation_time": 0.0012, "eval_runtime": 147.6519, "eval_samples_per_second": 16.539, "eval_steps_per_second": 8.269, "epoch": 0.0}, "model_type": "default", "objective": "LM loss", "add_l2r_token": true, "l2r_percent": 100.0, "embed_window_size": 1, "embed_loss_func": "cosine", "logical_per_device_eval_batch_size": 512, "per_device_eval_micro_batch_size": 2, "eval_gradient_accumulation_steps": 256, "world_size_env": 1, "world_size_runtime": 1, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-4968.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/checkpoint-4968",
|
| 4 |
+
"checkpoint": "checkpoint-4968",
|
| 5 |
+
"global_step": 4968,
|
| 6 |
+
"eval_loss": 3.8015518188476562,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.8015518188476562,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 147.4651,
|
| 11 |
+
"eval_samples_per_second": 16.56,
|
| 12 |
+
"eval_steps_per_second": 8.28,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": true,
|
| 34 |
+
"max_next_i_train": 5
|
| 35 |
+
}
|
results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-5040.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/checkpoint-5040",
|
| 4 |
+
"checkpoint": "checkpoint-5040",
|
| 5 |
+
"global_step": 5040,
|
| 6 |
+
"eval_loss": 3.7995407581329346,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.7995407581329346,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.0082,
|
| 11 |
+
"eval_samples_per_second": 16.725,
|
| 12 |
+
"eval_steps_per_second": 8.363,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": true,
|
| 34 |
+
"max_next_i_train": 5
|
| 35 |
+
}
|
results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-5112.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/checkpoint-5112",
|
| 4 |
+
"checkpoint": "checkpoint-5112",
|
| 5 |
+
"global_step": 5112,
|
| 6 |
+
"eval_loss": 3.797999858856201,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.797999858856201,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.2836,
|
| 11 |
+
"eval_samples_per_second": 16.694,
|
| 12 |
+
"eval_steps_per_second": 8.347,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": true,
|
| 34 |
+
"max_next_i_train": 5
|
| 35 |
+
}
|
results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-5184.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/checkpoint-5184",
|
| 4 |
+
"checkpoint": "checkpoint-5184",
|
| 5 |
+
"global_step": 5184,
|
| 6 |
+
"eval_loss": 3.7967641353607178,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.7967641353607178,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 145.9931,
|
| 11 |
+
"eval_samples_per_second": 16.727,
|
| 12 |
+
"eval_steps_per_second": 8.363,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": true,
|
| 34 |
+
"max_next_i_train": 5
|
| 35 |
+
}
|
results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-5256.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/checkpoint-5256",
|
| 4 |
+
"checkpoint": "checkpoint-5256",
|
| 5 |
+
"global_step": 5256,
|
| 6 |
+
"eval_loss": 3.798771381378174,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.798771381378174,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.0859,
|
| 11 |
+
"eval_samples_per_second": 16.716,
|
| 12 |
+
"eval_steps_per_second": 8.358,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": true,
|
| 34 |
+
"max_next_i_train": 5
|
| 35 |
+
}
|
results/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/l2r_n1_eval/checkpoint-5328.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896",
|
| 3 |
+
"checkpoint_path": "/workspace/concept_model/runs/random5-l2r50-i5-exp-fulle-lm-wsd-from-4896/checkpoint-5328",
|
| 4 |
+
"checkpoint": "checkpoint-5328",
|
| 5 |
+
"global_step": 5328,
|
| 6 |
+
"eval_loss": 3.7920925617218018,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.7920925617218018,
|
| 9 |
+
"eval_model_preparation_time": 0.0012,
|
| 10 |
+
"eval_runtime": 146.175,
|
| 11 |
+
"eval_samples_per_second": 16.706,
|
| 12 |
+
"eval_steps_per_second": 8.353,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"model_type": "default",
|
| 16 |
+
"objective": "LM loss",
|
| 17 |
+
"add_l2r_token": true,
|
| 18 |
+
"l2r_percent": 100.0,
|
| 19 |
+
"embed_window_size": 1,
|
| 20 |
+
"embed_loss_func": "cosine",
|
| 21 |
+
"logical_per_device_eval_batch_size": 512,
|
| 22 |
+
"per_device_eval_micro_batch_size": 2,
|
| 23 |
+
"eval_gradient_accumulation_steps": 256,
|
| 24 |
+
"world_size_env": 1,
|
| 25 |
+
"world_size_runtime": 1,
|
| 26 |
+
"global_logical_eval_batch": 512,
|
| 27 |
+
"global_eval_batch_size_arg": 512,
|
| 28 |
+
"batch_size_per_device_arg": 64,
|
| 29 |
+
"dataloader_num_workers": 4,
|
| 30 |
+
"dataloader_prefetch_factor": 4,
|
| 31 |
+
"dataloader_pin_memory": true,
|
| 32 |
+
"remove_unused_columns": true,
|
| 33 |
+
"next_i_enabled": true,
|
| 34 |
+
"max_next_i_train": 5
|
| 35 |
+
}
|