| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 4.737985134124756, | |
| "learning_rate": 5e-06, | |
| "loss": 1.5912, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 7.149003982543945, | |
| "learning_rate": 1e-05, | |
| "loss": 3.2011, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 6.096139430999756, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.2445, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 6.4711103439331055, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4005, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 6.055766582489014, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.9311, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 23.06619644165039, | |
| "learning_rate": 3e-05, | |
| "loss": 2.2437, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 9.455103874206543, | |
| "learning_rate": 3.5e-05, | |
| "loss": 2.7137, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 8.64001750946045, | |
| "learning_rate": 4e-05, | |
| "loss": 2.6398, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 58.382110595703125, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.2014, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 5.963547706604004, | |
| "learning_rate": 5e-05, | |
| "loss": 2.0685, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 5.909413814544678, | |
| "learning_rate": 4.99847706754774e-05, | |
| "loss": 3.1558, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 4.110403537750244, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 2.2536, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 17.61237907409668, | |
| "learning_rate": 4.9863047384206835e-05, | |
| "loss": 3.2304, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 42.55618667602539, | |
| "learning_rate": 4.975670171853926e-05, | |
| "loss": 1.8168, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 7.033435344696045, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.7345, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 6.939785480499268, | |
| "learning_rate": 4.9453690018345144e-05, | |
| "loss": 1.0468, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 7.7263078689575195, | |
| "learning_rate": 4.925739315689991e-05, | |
| "loss": 1.6574, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 14.468883514404297, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.6383, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 5.9138946533203125, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 0.8235, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 11.40308666229248, | |
| "learning_rate": 4.849231551964771e-05, | |
| "loss": 0.489, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 20.220449447631836, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.4448, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 29.097707748413086, | |
| "learning_rate": 4.783863644106502e-05, | |
| "loss": 1.7983, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 16.577442169189453, | |
| "learning_rate": 4.7469851157479177e-05, | |
| "loss": 1.8216, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 3.516134262084961, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 2.0506, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 3.0684404373168945, | |
| "learning_rate": 4.665063509461097e-05, | |
| "loss": 1.258, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 4.914421558380127, | |
| "learning_rate": 4.620120240391065e-05, | |
| "loss": 0.3911, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 6.704916954040527, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.6957, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 5.172149181365967, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 1.3392, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 2.9674534797668457, | |
| "learning_rate": 4.4700268840168045e-05, | |
| "loss": 0.6846, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 2.472653388977051, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.2735, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 2.405486583709717, | |
| "learning_rate": 4.357862063693486e-05, | |
| "loss": 0.2887, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 2.07614803314209, | |
| "learning_rate": 4.2983495008466276e-05, | |
| "loss": 0.3026, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 7.051891803741455, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.688, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 3.9787468910217285, | |
| "learning_rate": 4.172826515897146e-05, | |
| "loss": 2.5623, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 5.217522621154785, | |
| "learning_rate": 4.1069690242163484e-05, | |
| "loss": 1.2036, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 5.582896709442139, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 1.3976, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 4.498598098754883, | |
| "learning_rate": 3.969463130731183e-05, | |
| "loss": 1.6016, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 3.115513324737549, | |
| "learning_rate": 3.897982258676867e-05, | |
| "loss": 0.7646, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 3.488847494125366, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.5091, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 5.372682094573975, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.3678, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 5.9022674560546875, | |
| "learning_rate": 3.673678906964727e-05, | |
| "loss": 2.1068, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 3.6478700637817383, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 1.4413, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 7.891127109527588, | |
| "learning_rate": 3.516841607689501e-05, | |
| "loss": 1.2757, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 5.28378438949585, | |
| "learning_rate": 3.436516483539781e-05, | |
| "loss": 0.7885, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 6.047487735748291, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 2.3279, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 2.6575264930725098, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 0.5281, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 2.3986458778381348, | |
| "learning_rate": 3.1890933895424976e-05, | |
| "loss": 0.2615, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.6229203939437866, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.4922, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 1.5451277494430542, | |
| "learning_rate": 3.0197792270443982e-05, | |
| "loss": 0.2118, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.311353087425232, | |
| "learning_rate": 2.9341204441673266e-05, | |
| "loss": 0.1915, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 1.5571472644805908, | |
| "eval_runtime": 5999.5519, | |
| "eval_samples_per_second": 5.286, | |
| "eval_steps_per_second": 5.286, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 1.2530894241792e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |