| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5010266940451745, | |
| "eval_steps": 4, | |
| "global_step": 61, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008213552361396304, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.4077, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03285420944558522, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.3585, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.03285420944558522, | |
| "eval_loss": 2.385610342025757, | |
| "eval_runtime": 1029.8195, | |
| "eval_samples_per_second": 1.68, | |
| "eval_steps_per_second": 0.027, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.06570841889117043, | |
| "grad_norm": 1.8214708747240431, | |
| "learning_rate": 1.8e-06, | |
| "loss": 2.3759, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.06570841889117043, | |
| "eval_loss": 2.3762412071228027, | |
| "eval_runtime": 1083.1444, | |
| "eval_samples_per_second": 1.597, | |
| "eval_steps_per_second": 0.026, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.09856262833675565, | |
| "grad_norm": 1.5766702181423196, | |
| "learning_rate": 2.9473214285714287e-06, | |
| "loss": 2.3831, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.09856262833675565, | |
| "eval_loss": 2.3503129482269287, | |
| "eval_runtime": 1053.8602, | |
| "eval_samples_per_second": 1.642, | |
| "eval_steps_per_second": 0.027, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.13141683778234087, | |
| "grad_norm": 1.2634291220024447, | |
| "learning_rate": 2.7366071428571428e-06, | |
| "loss": 2.3484, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.13141683778234087, | |
| "eval_loss": 2.328130006790161, | |
| "eval_runtime": 1079.2806, | |
| "eval_samples_per_second": 1.603, | |
| "eval_steps_per_second": 0.026, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.16427104722792607, | |
| "grad_norm": 0.9006871585538231, | |
| "learning_rate": 2.5785714285714286e-06, | |
| "loss": 2.284, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16427104722792607, | |
| "eval_loss": 2.3153390884399414, | |
| "eval_runtime": 1065.0764, | |
| "eval_samples_per_second": 1.624, | |
| "eval_steps_per_second": 0.026, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1971252566735113, | |
| "grad_norm": 1.2412345963907103, | |
| "learning_rate": 2.3678571428571427e-06, | |
| "loss": 2.3009, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1971252566735113, | |
| "eval_loss": 2.306180953979492, | |
| "eval_runtime": 1024.6423, | |
| "eval_samples_per_second": 1.688, | |
| "eval_steps_per_second": 0.027, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2299794661190965, | |
| "grad_norm": 2.0067919017684916, | |
| "learning_rate": 2.157142857142857e-06, | |
| "loss": 2.2753, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.2299794661190965, | |
| "eval_loss": 2.297666072845459, | |
| "eval_runtime": 1037.1557, | |
| "eval_samples_per_second": 1.668, | |
| "eval_steps_per_second": 0.027, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.26283367556468173, | |
| "grad_norm": 1.0255281862331636, | |
| "learning_rate": 1.999107142857143e-06, | |
| "loss": 2.2783, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.26283367556468173, | |
| "eval_loss": 2.2926008701324463, | |
| "eval_runtime": 1052.6611, | |
| "eval_samples_per_second": 1.643, | |
| "eval_steps_per_second": 0.027, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.29568788501026694, | |
| "grad_norm": 0.8102039462592776, | |
| "learning_rate": 1.7883928571428571e-06, | |
| "loss": 2.2743, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.29568788501026694, | |
| "eval_loss": 2.2882704734802246, | |
| "eval_runtime": 1036.8504, | |
| "eval_samples_per_second": 1.669, | |
| "eval_steps_per_second": 0.027, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.32854209445585214, | |
| "grad_norm": 0.6628155960720566, | |
| "learning_rate": 1.6303571428571428e-06, | |
| "loss": 2.2801, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.32854209445585214, | |
| "eval_loss": 2.2851009368896484, | |
| "eval_runtime": 1043.8103, | |
| "eval_samples_per_second": 1.657, | |
| "eval_steps_per_second": 0.027, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3613963039014374, | |
| "grad_norm": 0.674490895719983, | |
| "learning_rate": 1.4723214285714284e-06, | |
| "loss": 2.2776, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.3613963039014374, | |
| "eval_loss": 2.2824671268463135, | |
| "eval_runtime": 1068.5698, | |
| "eval_samples_per_second": 1.619, | |
| "eval_steps_per_second": 0.026, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.3942505133470226, | |
| "grad_norm": 0.6816127601956803, | |
| "learning_rate": 1.261607142857143e-06, | |
| "loss": 2.2633, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3942505133470226, | |
| "eval_loss": 2.2798171043395996, | |
| "eval_runtime": 1061.4491, | |
| "eval_samples_per_second": 1.63, | |
| "eval_steps_per_second": 0.026, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.4271047227926078, | |
| "grad_norm": 1.258863916463987, | |
| "learning_rate": 1.0508928571428572e-06, | |
| "loss": 2.2655, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.4271047227926078, | |
| "eval_loss": 2.2778077125549316, | |
| "eval_runtime": 1092.5301, | |
| "eval_samples_per_second": 1.583, | |
| "eval_steps_per_second": 0.026, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.459958932238193, | |
| "grad_norm": 0.798209284858603, | |
| "learning_rate": 8.401785714285717e-07, | |
| "loss": 2.2677, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.459958932238193, | |
| "eval_loss": 2.276536226272583, | |
| "eval_runtime": 1074.6473, | |
| "eval_samples_per_second": 1.61, | |
| "eval_steps_per_second": 0.026, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.4928131416837782, | |
| "grad_norm": 0.9625257358273914, | |
| "learning_rate": 6.821428571428571e-07, | |
| "loss": 2.2702, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4928131416837782, | |
| "eval_loss": 2.275395154953003, | |
| "eval_runtime": 1074.6135, | |
| "eval_samples_per_second": 1.61, | |
| "eval_steps_per_second": 0.026, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5010266940451745, | |
| "step": 61, | |
| "total_flos": 240638427660288.0, | |
| "train_loss": 2.305594545895936, | |
| "train_runtime": 34982.9446, | |
| "train_samples_per_second": 0.222, | |
| "train_steps_per_second": 0.002 | |
| } | |
| ], | |
| "logging_steps": 4, | |
| "max_steps": 61, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "total_flos": 240638427660288.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |