Training in progress, step 150

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b41a22bf309b8629c877144eef78fcc4225ffe8d1c84eb50c2e15b24fa71d05
+oid sha256:b2ef83fb36d59272e72cfa6efb6c2cf0cc15bfd49a4c87828ca8cf8c1563f3c7
 size 6627156248

logging.jsonl

@@ -152,3 +152,14 @@
 {"loss": 0.4786377, "grad_norm": 0.0880352, "learning_rate": 3.412e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.004046, "rewards/chosen": -0.03485107, "rewards/rejected": -0.27191162, "rewards/accuracies": 0.75, "rewards/margins": 0.2366333, "logps/rejected": -2.72070312, "logps/chosen": -0.34814453, "logits/rejected": -0.33129883, "logits/chosen": -0.56591797, "nll_loss": 0.34790039, "log_odds_ratio": -0.35011292, "log_odds_chosen": 3.12890625, "epoch": 0.32552693, "global_step/max_steps": "139/427", "percentage": "32.55%", "elapsed_time": "9h 32m 2s", "remaining_time": "19h 45m 14s"}
 {"loss": 0.42858887, "grad_norm": 0.14312401, "learning_rate": 3.4e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.004058, "rewards/chosen": -0.03723145, "rewards/rejected": -0.08087158, "rewards/accuracies": 0.75, "rewards/margins": 0.0435791, "logps/rejected": -0.80957031, "logps/chosen": -0.37255859, "logits/rejected": -0.28607178, "logits/chosen": -0.56787109, "nll_loss": 0.37255859, "log_odds_ratio": -0.44506836, "log_odds_chosen": 0.81738281, "epoch": 0.32786885, "global_step/max_steps": "140/427", "percentage": "32.79%", "elapsed_time": "9h 34m 27s", "remaining_time": "19h 37m 38s"}
 {"eval_loss": 0.46778965, "eval_runtime": 649.5435, "eval_samples_per_second": 0.65, "eval_steps_per_second": 0.082, "eval_rewards/chosen": -0.04254496, "eval_rewards/rejected": -0.23106067, "eval_rewards/accuracies": 0.6981132, "eval_rewards/margins": 0.18846275, "eval_logps/rejected": -2.31296062, "eval_logps/chosen": -0.42576283, "eval_logits/rejected": -0.31001714, "eval_logits/chosen": -0.84275502, "eval_nll_loss": 0.42568913, "eval_log_odds_ratio": -0.44461232, "eval_log_odds_chosen": 2.32117844, "epoch": 0.32786885, "global_step/max_steps": "140/427", "percentage": "32.79%", "elapsed_time": "9h 45m 16s", "remaining_time": "19h 59m 49s"}
+{"loss": 0.47134399, "grad_norm": 0.09910604, "learning_rate": 3.389e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.003978, "rewards/chosen": -0.03421021, "rewards/rejected": -0.11875916, "rewards/accuracies": 0.75, "rewards/margins": 0.08465576, "logps/rejected": -1.18701172, "logps/chosen": -0.34204102, "logits/rejected": -0.40197754, "logits/chosen": -0.74609375, "nll_loss": 0.34179688, "log_odds_ratio": -0.51373291, "log_odds_chosen": 1.18334961, "epoch": 0.33021077, "global_step/max_steps": "141/427", "percentage": "33.02%", "elapsed_time": "9h 50m 9s", "remaining_time": "19h 57m 4s"}
+{"loss": 0.44058228, "grad_norm": 0.09278602, "learning_rate": 3.377e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.003981, "rewards/chosen": -0.04788208, "rewards/rejected": -0.10705566, "rewards/accuracies": 0.5, "rewards/margins": 0.05908203, "logps/rejected": -1.0703125, "logps/chosen": -0.47851562, "logits/rejected": -0.53686523, "logits/chosen": -1.09960938, "nll_loss": 0.47900391, "log_odds_ratio": -0.51629639, "log_odds_chosen": 0.91894531, "epoch": 0.33255269, "global_step/max_steps": "142/427", "percentage": "33.26%", "elapsed_time": "9h 53m 51s", "remaining_time": "19h 51m 53s"}
+{"loss": 0.44866943, "grad_norm": 0.10309532, "learning_rate": 3.365e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.003986, "rewards/chosen": -0.04376221, "rewards/rejected": -0.3314209, "rewards/accuracies": 0.75, "rewards/margins": 0.28765869, "logps/rejected": -3.31689453, "logps/chosen": -0.4375, "logits/rejected": -0.30639648, "logits/chosen": -0.83032227, "nll_loss": 0.4375, "log_odds_ratio": -0.27932739, "log_odds_chosen": 3.61914062, "epoch": 0.33489461, "global_step/max_steps": "143/427", "percentage": "33.49%", "elapsed_time": "9h 57m 19s", "remaining_time": "19h 46m 18s"}
+{"loss": 0.42626953, "grad_norm": 0.08816983, "learning_rate": 3.353e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.003999, "rewards/chosen": -0.04168701, "rewards/rejected": -0.46606445, "rewards/accuracies": 1.0, "rewards/margins": 0.42456055, "logps/rejected": -4.65576172, "logps/chosen": -0.41699219, "logits/rejected": -0.12660217, "logits/chosen": -0.90576172, "nll_loss": 0.41748047, "log_odds_ratio": -0.15431213, "log_odds_chosen": 5.02001953, "epoch": 0.33723653, "global_step/max_steps": "144/427", "percentage": "33.72%", "elapsed_time": "9h 59m 29s", "remaining_time": "19h 38m 10s"}
+{"loss": 0.72900391, "grad_norm": 0.8846494, "learning_rate": 3.341e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.004012, "rewards/chosen": -0.0458374, "rewards/rejected": -0.18359375, "rewards/accuracies": 0.5, "rewards/margins": 0.13793945, "logps/rejected": -1.83203125, "logps/chosen": -0.45849609, "logits/rejected": -0.38061523, "logits/chosen": -0.68237305, "nll_loss": 0.45849609, "log_odds_ratio": -0.78320312, "log_odds_chosen": 1.34521484, "epoch": 0.33957845, "global_step/max_steps": "145/427", "percentage": "33.96%", "elapsed_time": "10h 1m 47s", "remaining_time": "19h 30m 23s"}
+{"loss": 0.43817139, "grad_norm": 0.14548092, "learning_rate": 3.329e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.004026, "rewards/chosen": -0.03140259, "rewards/rejected": -0.06945801, "rewards/accuracies": 1.0, "rewards/margins": 0.03805542, "logps/rejected": -0.69384766, "logps/chosen": -0.31396484, "logits/rejected": -0.24462891, "logits/chosen": -0.93530273, "nll_loss": 0.31445312, "log_odds_ratio": -0.39233398, "log_odds_chosen": 0.89501953, "epoch": 0.34192037, "global_step/max_steps": "146/427", "percentage": "34.19%", "elapsed_time": "10h 3m 47s", "remaining_time": "19h 22m 6s"}
+{"loss": 0.44674683, "grad_norm": 0.10295217, "learning_rate": 3.318e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.004028, "rewards/chosen": -0.04327393, "rewards/rejected": -0.05230713, "rewards/accuracies": 1.0, "rewards/margins": 0.0090332, "logps/rejected": -0.52294922, "logps/chosen": -0.43310547, "logits/rejected": -0.28613281, "logits/chosen": -1.01757812, "nll_loss": 0.43310547, "log_odds_ratio": -0.578125, "log_odds_chosen": 0.25488281, "epoch": 0.3442623, "global_step/max_steps": "147/427", "percentage": "34.43%", "elapsed_time": "10h 7m 40s", "remaining_time": "19h 17m 29s"}
+{"loss": 0.43991089, "grad_norm": 0.09682243, "learning_rate": 3.306e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.004035, "rewards/chosen": -0.03491211, "rewards/rejected": -0.03268433, "rewards/accuracies": 0.5, "rewards/margins": -0.00222778, "logps/rejected": -0.32714844, "logps/chosen": -0.34912109, "logits/rejected": -0.15167236, "logits/chosen": -0.96386719, "nll_loss": 0.34912109, "log_odds_ratio": -0.71582031, "log_odds_chosen": -0.02294922, "epoch": 0.34660422, "global_step/max_steps": "148/427", "percentage": "34.66%", "elapsed_time": "10h 10m 44s", "remaining_time": "19h 11m 19s"}
+{"loss": 0.47558594, "grad_norm": 0.10571572, "learning_rate": 3.294e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.004039, "rewards/chosen": -0.03918457, "rewards/rejected": -0.21398926, "rewards/accuracies": 0.25, "rewards/margins": 0.17510986, "logps/rejected": -2.13623047, "logps/chosen": -0.39160156, "logits/rejected": -0.4453125, "logits/chosen": -0.48193359, "nll_loss": 0.39160156, "log_odds_ratio": -0.70800781, "log_odds_chosen": 1.984375, "epoch": 0.34894614, "global_step/max_steps": "149/427", "percentage": "34.89%", "elapsed_time": "10h 14m 10s", "remaining_time": "19h 5m 55s"}
+{"loss": 0.40313721, "grad_norm": 0.11789128, "learning_rate": 3.282e-05, "memory(GiB)": 133.71, "train_speed(iter/s)": 0.004048, "rewards/chosen": -0.02890015, "rewards/rejected": -0.21908569, "rewards/accuracies": 1.0, "rewards/margins": 0.18978882, "logps/rejected": -2.19287109, "logps/chosen": -0.2890625, "logits/rejected": -0.1060791, "logits/chosen": -0.59295654, "nll_loss": 0.28930664, "log_odds_ratio": -0.47265625, "log_odds_chosen": 2.34716797, "epoch": 0.35128806, "global_step/max_steps": "150/427", "percentage": "35.13%", "elapsed_time": "10h 16m 56s", "remaining_time": "18h 59m 17s"}
+{"eval_loss": 0.47329032, "eval_runtime": 649.4744, "eval_samples_per_second": 0.65, "eval_steps_per_second": 0.082, "eval_rewards/chosen": -0.04066784, "eval_rewards/rejected": -0.09058308, "eval_rewards/accuracies": 0.56603771, "eval_rewards/margins": 0.04997282, "eval_logps/rejected": -0.90518129, "eval_logps/chosen": -0.40683961, "eval_logits/rejected": -0.22765307, "eval_logits/chosen": -0.72029084, "eval_nll_loss": 0.40660009, "eval_log_odds_ratio": -0.68672609, "eval_log_odds_chosen": 0.5650059, "epoch": 0.35128806, "global_step/max_steps": "150/427", "percentage": "35.13%", "elapsed_time": "10h 27m 46s", "remaining_time": "19h 19m 16s"}