| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5370857725978839, | |
| "global_step": 20000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.973145711370106e-05, | |
| "loss": 0.2809, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 0.9594895839691162, | |
| "eval_runtime": 2.5122, | |
| "eval_samples_per_second": 1378.065, | |
| "eval_steps_per_second": 7.563, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.946291422740212e-05, | |
| "loss": 0.2549, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 1.03871750831604, | |
| "eval_runtime": 2.5032, | |
| "eval_samples_per_second": 1383.049, | |
| "eval_steps_per_second": 7.59, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.9194371341103174e-05, | |
| "loss": 0.2473, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 0.969585120677948, | |
| "eval_runtime": 2.5286, | |
| "eval_samples_per_second": 1369.122, | |
| "eval_steps_per_second": 7.514, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.8925828454804235e-05, | |
| "loss": 0.2514, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.8840947151184082, | |
| "eval_runtime": 2.5014, | |
| "eval_samples_per_second": 1384.048, | |
| "eval_steps_per_second": 7.596, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.8657285568505293e-05, | |
| "loss": 0.2481, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 0.9262232780456543, | |
| "eval_runtime": 2.509, | |
| "eval_samples_per_second": 1379.819, | |
| "eval_steps_per_second": 7.573, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.838874268220635e-05, | |
| "loss": 0.2428, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 0.9006544947624207, | |
| "eval_runtime": 2.5007, | |
| "eval_samples_per_second": 1384.421, | |
| "eval_steps_per_second": 7.598, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.8120199795907408e-05, | |
| "loss": 0.2411, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 0.9295358657836914, | |
| "eval_runtime": 2.5648, | |
| "eval_samples_per_second": 1349.806, | |
| "eval_steps_per_second": 7.408, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.7851656909608465e-05, | |
| "loss": 0.2393, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_loss": 0.8647714853286743, | |
| "eval_runtime": 2.5012, | |
| "eval_samples_per_second": 1384.115, | |
| "eval_steps_per_second": 7.596, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.7583114023309523e-05, | |
| "loss": 0.2389, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 0.8838908672332764, | |
| "eval_runtime": 2.5044, | |
| "eval_samples_per_second": 1382.376, | |
| "eval_steps_per_second": 7.587, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.7314571137010584e-05, | |
| "loss": 0.2402, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 0.927254319190979, | |
| "eval_runtime": 2.5033, | |
| "eval_samples_per_second": 1382.963, | |
| "eval_steps_per_second": 7.59, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.7046028250711638e-05, | |
| "loss": 0.2342, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.828113317489624, | |
| "eval_runtime": 2.503, | |
| "eval_samples_per_second": 1383.137, | |
| "eval_steps_per_second": 7.591, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.67774853644127e-05, | |
| "loss": 0.2336, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 0.9690735936164856, | |
| "eval_runtime": 2.5071, | |
| "eval_samples_per_second": 1380.884, | |
| "eval_steps_per_second": 7.579, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.6508942478113756e-05, | |
| "loss": 0.2363, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 0.8396281003952026, | |
| "eval_runtime": 2.5067, | |
| "eval_samples_per_second": 1381.116, | |
| "eval_steps_per_second": 7.58, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.6240399591814814e-05, | |
| "loss": 0.2337, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 0.8618175387382507, | |
| "eval_runtime": 2.5018, | |
| "eval_samples_per_second": 1383.782, | |
| "eval_steps_per_second": 7.594, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.597185670551587e-05, | |
| "loss": 0.2307, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.9263376593589783, | |
| "eval_runtime": 2.5033, | |
| "eval_samples_per_second": 1382.991, | |
| "eval_steps_per_second": 7.59, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.570331381921693e-05, | |
| "loss": 0.2347, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 0.8525048494338989, | |
| "eval_runtime": 2.5022, | |
| "eval_samples_per_second": 1383.595, | |
| "eval_steps_per_second": 7.593, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.543477093291799e-05, | |
| "loss": 0.2294, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 0.9026502966880798, | |
| "eval_runtime": 2.503, | |
| "eval_samples_per_second": 1383.13, | |
| "eval_steps_per_second": 7.591, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.5166228046619047e-05, | |
| "loss": 0.2297, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 0.8343677520751953, | |
| "eval_runtime": 2.503, | |
| "eval_samples_per_second": 1383.136, | |
| "eval_steps_per_second": 7.591, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.4897685160320103e-05, | |
| "loss": 0.2294, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 0.8918474912643433, | |
| "eval_runtime": 2.5026, | |
| "eval_samples_per_second": 1383.364, | |
| "eval_steps_per_second": 7.592, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1.4629142274021162e-05, | |
| "loss": 0.2322, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 0.832206130027771, | |
| "eval_runtime": 2.5024, | |
| "eval_samples_per_second": 1383.452, | |
| "eval_steps_per_second": 7.593, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.436059938772222e-05, | |
| "loss": 0.2327, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 0.8489925861358643, | |
| "eval_runtime": 2.5027, | |
| "eval_samples_per_second": 1383.288, | |
| "eval_steps_per_second": 7.592, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1.4092056501423279e-05, | |
| "loss": 0.2313, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.8496459722518921, | |
| "eval_runtime": 2.4999, | |
| "eval_samples_per_second": 1384.839, | |
| "eval_steps_per_second": 7.6, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.3823513615124335e-05, | |
| "loss": 0.2305, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.9429409503936768, | |
| "eval_runtime": 2.5651, | |
| "eval_samples_per_second": 1349.664, | |
| "eval_steps_per_second": 7.407, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.3554970728825394e-05, | |
| "loss": 0.2304, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 0.9687642455101013, | |
| "eval_runtime": 2.497, | |
| "eval_samples_per_second": 1386.456, | |
| "eval_steps_per_second": 7.609, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.3286427842526453e-05, | |
| "loss": 0.2295, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_loss": 0.9534400701522827, | |
| "eval_runtime": 2.5012, | |
| "eval_samples_per_second": 1384.153, | |
| "eval_steps_per_second": 7.596, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 1.3017884956227511e-05, | |
| "loss": 0.2328, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.9109036326408386, | |
| "eval_runtime": 2.5038, | |
| "eval_samples_per_second": 1382.713, | |
| "eval_steps_per_second": 7.589, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.274934206992857e-05, | |
| "loss": 0.2269, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 0.9277821779251099, | |
| "eval_runtime": 2.5098, | |
| "eval_samples_per_second": 1379.412, | |
| "eval_steps_per_second": 7.57, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.2480799183629626e-05, | |
| "loss": 0.2247, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 0.8515273332595825, | |
| "eval_runtime": 2.5018, | |
| "eval_samples_per_second": 1383.826, | |
| "eval_steps_per_second": 7.595, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.2212256297330685e-05, | |
| "loss": 0.2233, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_loss": 0.8787582516670227, | |
| "eval_runtime": 2.5026, | |
| "eval_samples_per_second": 1383.344, | |
| "eval_steps_per_second": 7.592, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1.1943713411031743e-05, | |
| "loss": 0.2257, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.9345625638961792, | |
| "eval_runtime": 2.502, | |
| "eval_samples_per_second": 1383.68, | |
| "eval_steps_per_second": 7.594, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1.1675170524732802e-05, | |
| "loss": 0.2263, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 0.9080494046211243, | |
| "eval_runtime": 2.5009, | |
| "eval_samples_per_second": 1384.285, | |
| "eval_steps_per_second": 7.597, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1.1406627638433858e-05, | |
| "loss": 0.2254, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_loss": 0.9714885950088501, | |
| "eval_runtime": 2.5027, | |
| "eval_samples_per_second": 1383.281, | |
| "eval_steps_per_second": 7.592, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.1138084752134917e-05, | |
| "loss": 0.2228, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 0.9011908769607544, | |
| "eval_runtime": 2.4967, | |
| "eval_samples_per_second": 1386.629, | |
| "eval_steps_per_second": 7.61, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.0869541865835975e-05, | |
| "loss": 0.2234, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_loss": 0.8457213044166565, | |
| "eval_runtime": 2.5001, | |
| "eval_samples_per_second": 1384.748, | |
| "eval_steps_per_second": 7.6, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.0600998979537034e-05, | |
| "loss": 0.221, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 0.9264540672302246, | |
| "eval_runtime": 2.4947, | |
| "eval_samples_per_second": 1387.758, | |
| "eval_steps_per_second": 7.616, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.033245609323809e-05, | |
| "loss": 0.2229, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 0.9298267364501953, | |
| "eval_runtime": 2.4937, | |
| "eval_samples_per_second": 1388.279, | |
| "eval_steps_per_second": 7.619, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.0063913206939149e-05, | |
| "loss": 0.2245, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.9173328280448914, | |
| "eval_runtime": 2.5, | |
| "eval_samples_per_second": 1384.796, | |
| "eval_steps_per_second": 7.6, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 9.795370320640208e-06, | |
| "loss": 0.2232, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.9208750128746033, | |
| "eval_runtime": 2.4954, | |
| "eval_samples_per_second": 1387.373, | |
| "eval_steps_per_second": 7.614, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 9.526827434341266e-06, | |
| "loss": 0.2224, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_loss": 0.9066223502159119, | |
| "eval_runtime": 2.502, | |
| "eval_samples_per_second": 1383.684, | |
| "eval_steps_per_second": 7.594, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 9.258284548042323e-06, | |
| "loss": 0.2223, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_loss": 0.8743013739585876, | |
| "eval_runtime": 2.5598, | |
| "eval_samples_per_second": 1352.458, | |
| "eval_steps_per_second": 7.423, | |
| "step": 20000 | |
| } | |
| ], | |
| "max_steps": 37238, | |
| "num_train_epochs": 1, | |
| "total_flos": 2.8044971311104e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |