Training in progress, step 130

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f25463bb5f45f6f9bc1c30182e3e84c708fa721d54179e8a412b2f959349fdf6
+oid sha256:e2f12152976e2acde537479871935ba50770985026b0340f03ee13cdb679b503
 size 6627156248

logging.jsonl

@@ -130,3 +130,14 @@
 {"loss": 0.37322998, "grad_norm": 0.11315849, "learning_rate": 4.777e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003303, "rewards/chosen": -0.02914429, "rewards/rejected": -0.06541443, "rewards/accuracies": 0.25, "rewards/margins": 0.03622437, "logps/rejected": -0.65368652, "logps/chosen": -0.29174805, "logits/rejected": -0.4230957, "logits/chosen": -0.60546875, "nll_loss": 0.29223633, "log_odds_ratio": -0.75622559, "log_odds_chosen": 0.35253906, "epoch": 0.42198582, "global_step/max_steps": "119/846", "percentage": "14.07%", "elapsed_time": "10h 0m 6s", "remaining_time": "2d 13h 6m 14s"}
 {"loss": 0.37789917, "grad_norm": 0.1166358, "learning_rate": 4.77e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003305, "rewards/chosen": -0.04443359, "rewards/rejected": -0.11999512, "rewards/accuracies": 1.0, "rewards/margins": 0.07550049, "logps/rejected": -1.20019531, "logps/chosen": -0.4440918, "logits/rejected": -0.24209595, "logits/chosen": -0.18579102, "nll_loss": 0.4440918, "log_odds_ratio": -0.31967163, "log_odds_chosen": 1.44921875, "epoch": 0.42553191, "global_step/max_steps": "120/846", "percentage": "14.18%", "elapsed_time": "10h 4m 53s", "remaining_time": "2d 12h 59m 36s"}
 {"eval_loss": 0.41283742, "eval_runtime": 568.1176, "eval_samples_per_second": 0.491, "eval_steps_per_second": 0.062, "eval_rewards/chosen": -0.03860386, "eval_rewards/rejected": -0.12836914, "eval_rewards/accuracies": 0.65714288, "eval_rewards/margins": 0.08977748, "eval_logps/rejected": -1.28404021, "eval_logps/chosen": -0.38635603, "eval_logits/rejected": -0.39749581, "eval_logits/chosen": -0.51142579, "eval_nll_loss": 0.38696986, "eval_log_odds_ratio": -0.48230678, "eval_log_odds_chosen": 1.40002787, "epoch": 0.42553191, "global_step/max_steps": "120/846", "percentage": "14.18%", "elapsed_time": "10h 14m 21s", "remaining_time": "2d 13h 56m 53s"}
+{"loss": 0.48397827, "grad_norm": 0.21679665, "learning_rate": 4.763e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003256, "rewards/chosen": -0.05886841, "rewards/rejected": -0.1071167, "rewards/accuracies": 0.75, "rewards/margins": 0.04824829, "logps/rejected": -1.07128906, "logps/chosen": -0.58837891, "logits/rejected": -0.21166992, "logits/chosen": -0.40551758, "nll_loss": 0.58740234, "log_odds_ratio": -0.41650391, "log_odds_chosen": 0.80371094, "epoch": 0.42907801, "global_step/max_steps": "121/846", "percentage": "14.30%", "elapsed_time": "10h 19m 8s", "remaining_time": "2d 13h 49m 43s"}
+{"loss": 0.40228271, "grad_norm": 0.12474387, "learning_rate": 4.757e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003267, "rewards/chosen": -0.03393555, "rewards/rejected": -0.14855957, "rewards/accuracies": 1.0, "rewards/margins": 0.11450195, "logps/rejected": -1.48730469, "logps/chosen": -0.33911133, "logits/rejected": -0.69921875, "logits/chosen": -0.65087891, "nll_loss": 0.33935547, "log_odds_ratio": -0.19784546, "log_odds_chosen": 1.97412109, "epoch": 0.43262411, "global_step/max_steps": "122/846", "percentage": "14.42%", "elapsed_time": "10h 22m 5s", "remaining_time": "2d 13h 31m 46s"}
+{"loss": 0.35971069, "grad_norm": 0.14296877, "learning_rate": 4.75e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003272, "rewards/chosen": -0.03463745, "rewards/rejected": -0.12670898, "rewards/accuracies": 0.75, "rewards/margins": 0.09222412, "logps/rejected": -1.26757812, "logps/chosen": -0.34570312, "logits/rejected": -0.37158203, "logits/chosen": -0.47631836, "nll_loss": 0.34570312, "log_odds_ratio": -0.29193115, "log_odds_chosen": 1.56689453, "epoch": 0.43617021, "global_step/max_steps": "123/846", "percentage": "14.54%", "elapsed_time": "10h 26m 13s", "remaining_time": "2d 13h 20m 57s"}
+{"loss": 0.41882324, "grad_norm": 0.52877933, "learning_rate": 4.744e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003279, "rewards/chosen": -0.0362854, "rewards/rejected": -0.11141968, "rewards/accuracies": 0.75, "rewards/margins": 0.07507324, "logps/rejected": -1.11401367, "logps/chosen": -0.36279297, "logits/rejected": -0.37915039, "logits/chosen": -0.51855469, "nll_loss": 0.36328125, "log_odds_ratio": -0.57556152, "log_odds_chosen": 1.17773438, "epoch": 0.43971631, "global_step/max_steps": "124/846", "percentage": "14.66%", "elapsed_time": "10h 30m 1s", "remaining_time": "2d 13h 8m 22s"}
+{"loss": 0.31719971, "grad_norm": 0.09337451, "learning_rate": 4.737e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.00329, "rewards/chosen": -0.02603149, "rewards/rejected": -0.12472534, "rewards/accuracies": 0.75, "rewards/margins": 0.09890747, "logps/rejected": -1.24829102, "logps/chosen": -0.25976562, "logits/rejected": -0.46118164, "logits/chosen": -0.59179688, "nll_loss": 0.26000977, "log_odds_ratio": -0.50631714, "log_odds_chosen": 1.54296875, "epoch": 0.44326241, "global_step/max_steps": "125/846", "percentage": "14.78%", "elapsed_time": "10h 33m 1s", "remaining_time": "2d 12h 51m 14s"}
+{"loss": 0.41064453, "grad_norm": 0.16397984, "learning_rate": 4.731e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003293, "rewards/chosen": -0.06164551, "rewards/rejected": -0.4551239, "rewards/accuracies": 0.75, "rewards/margins": 0.39294434, "logps/rejected": -4.54248047, "logps/chosen": -0.61645508, "logits/rejected": -0.34417725, "logits/chosen": -0.4705658, "nll_loss": 0.61450195, "log_odds_ratio": -0.52001953, "log_odds_chosen": 4.28515625, "epoch": 0.44680851, "global_step/max_steps": "126/846", "percentage": "14.89%", "elapsed_time": "10h 37m 31s", "remaining_time": "2d 12h 43m 1s"}
+{"loss": 0.41320801, "grad_norm": 0.1379481, "learning_rate": 4.724e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003291, "rewards/chosen": -0.03250122, "rewards/rejected": -0.15670776, "rewards/accuracies": 0.5, "rewards/margins": 0.12405396, "logps/rejected": -1.56494141, "logps/chosen": -0.32446289, "logits/rejected": -0.36669922, "logits/chosen": -0.63317871, "nll_loss": 0.32495117, "log_odds_ratio": -0.48278809, "log_odds_chosen": 1.78759766, "epoch": 0.45035461, "global_step/max_steps": "127/846", "percentage": "15.01%", "elapsed_time": "10h 42m 50s", "remaining_time": "2d 12h 39m 20s"}
+{"loss": 0.3855896, "grad_norm": 0.20181563, "learning_rate": 4.717e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003298, "rewards/chosen": -0.02474976, "rewards/rejected": -0.09185791, "rewards/accuracies": 1.0, "rewards/margins": 0.06707764, "logps/rejected": -0.91845703, "logps/chosen": -0.24755859, "logits/rejected": -0.37646484, "logits/chosen": -0.39111328, "nll_loss": 0.24682617, "log_odds_ratio": -0.22546387, "log_odds_chosen": 1.60351562, "epoch": 0.45390071, "global_step/max_steps": "128/846", "percentage": "15.13%", "elapsed_time": "10h 46m 33s", "remaining_time": "2d 12h 26m 45s"}
+{"loss": 0.36123657, "grad_norm": 0.19869718, "learning_rate": 4.711e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003305, "rewards/chosen": -0.03344727, "rewards/rejected": -0.08837891, "rewards/accuracies": 0.5, "rewards/margins": 0.05499268, "logps/rejected": -0.88378906, "logps/chosen": -0.33422852, "logits/rejected": -0.38818359, "logits/chosen": -0.60449219, "nll_loss": 0.33422852, "log_odds_ratio": -0.68121338, "log_odds_chosen": 0.76708984, "epoch": 0.45744681, "global_step/max_steps": "129/846", "percentage": "15.25%", "elapsed_time": "10h 50m 17s", "remaining_time": "2d 12h 14m 22s"}
+{"loss": 0.37228394, "grad_norm": 0.11339411, "learning_rate": 4.704e-05, "memory(GiB)": 132.99, "train_speed(iter/s)": 0.003311, "rewards/chosen": -0.03164673, "rewards/rejected": -0.03961182, "rewards/accuracies": 0.75, "rewards/margins": 0.00796509, "logps/rejected": -0.39624023, "logps/chosen": -0.31665039, "logits/rejected": -0.38671875, "logits/chosen": -0.43017578, "nll_loss": 0.31665039, "log_odds_ratio": -0.60546875, "log_odds_chosen": 0.26025391, "epoch": 0.46099291, "global_step/max_steps": "130/846", "percentage": "15.37%", "elapsed_time": "10h 54m 11s", "remaining_time": "2d 12h 3m 7s"}
+{"eval_loss": 0.38319752, "eval_runtime": 568.3988, "eval_samples_per_second": 0.491, "eval_steps_per_second": 0.062, "eval_rewards/chosen": -0.03471157, "eval_rewards/rejected": -0.34204799, "eval_rewards/accuracies": 0.77142859, "eval_rewards/margins": 0.30736259, "eval_logps/rejected": -3.4203124, "eval_logps/chosen": -0.34718192, "eval_logits/rejected": -0.40456194, "eval_logits/chosen": -0.52040666, "eval_nll_loss": 0.34737724, "eval_log_odds_ratio": -0.39023611, "eval_log_odds_chosen": 3.7414341, "epoch": 0.46099291, "global_step/max_steps": "130/846", "percentage": "15.37%", "elapsed_time": "11h 3m 40s", "remaining_time": "2d 12h 55m 18s"}