| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.5, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.18797126412391663, | |
| "learning_rate": 9.949748743718594e-05, | |
| "loss": 0.6604, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.15986427664756775, | |
| "learning_rate": 9.84924623115578e-05, | |
| "loss": 0.6109, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.14691925048828125, | |
| "learning_rate": 9.748743718592965e-05, | |
| "loss": 0.6119, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.12482919543981552, | |
| "learning_rate": 9.64824120603015e-05, | |
| "loss": 0.6518, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.1446041613817215, | |
| "learning_rate": 9.547738693467337e-05, | |
| "loss": 0.5964, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.11769542098045349, | |
| "learning_rate": 9.447236180904523e-05, | |
| "loss": 0.5714, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.1492375284433365, | |
| "learning_rate": 9.34673366834171e-05, | |
| "loss": 0.6074, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.16033588349819183, | |
| "learning_rate": 9.246231155778895e-05, | |
| "loss": 0.5441, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.12141509354114532, | |
| "learning_rate": 9.14572864321608e-05, | |
| "loss": 0.5898, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.14080357551574707, | |
| "learning_rate": 9.045226130653267e-05, | |
| "loss": 0.5599, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.14213934540748596, | |
| "learning_rate": 8.944723618090453e-05, | |
| "loss": 0.5713, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.1625017374753952, | |
| "learning_rate": 8.84422110552764e-05, | |
| "loss": 0.543, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.16236303746700287, | |
| "learning_rate": 8.743718592964825e-05, | |
| "loss": 0.5461, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.14767853915691376, | |
| "learning_rate": 8.64321608040201e-05, | |
| "loss": 0.615, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.1415889859199524, | |
| "learning_rate": 8.542713567839196e-05, | |
| "loss": 0.5717, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.14877167344093323, | |
| "learning_rate": 8.442211055276383e-05, | |
| "loss": 0.5413, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.15324772894382477, | |
| "learning_rate": 8.341708542713568e-05, | |
| "loss": 0.5564, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.14583168923854828, | |
| "learning_rate": 8.241206030150754e-05, | |
| "loss": 0.549, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.17882034182548523, | |
| "learning_rate": 8.14070351758794e-05, | |
| "loss": 0.5495, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.13743259012699127, | |
| "learning_rate": 8.040201005025126e-05, | |
| "loss": 0.5545, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 0.1426534354686737, | |
| "learning_rate": 7.939698492462313e-05, | |
| "loss": 0.5137, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.16195817291736603, | |
| "learning_rate": 7.839195979899498e-05, | |
| "loss": 0.5241, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 0.17723768949508667, | |
| "learning_rate": 7.738693467336684e-05, | |
| "loss": 0.5201, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.1717960685491562, | |
| "learning_rate": 7.638190954773869e-05, | |
| "loss": 0.4599, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.17444512248039246, | |
| "learning_rate": 7.537688442211056e-05, | |
| "loss": 0.5151, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.19015884399414062, | |
| "learning_rate": 7.437185929648241e-05, | |
| "loss": 0.5413, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.1704825609922409, | |
| "learning_rate": 7.336683417085427e-05, | |
| "loss": 0.5352, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.18583402037620544, | |
| "learning_rate": 7.236180904522614e-05, | |
| "loss": 0.4971, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.1627408117055893, | |
| "learning_rate": 7.135678391959799e-05, | |
| "loss": 0.4678, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.167761892080307, | |
| "learning_rate": 7.035175879396985e-05, | |
| "loss": 0.5322, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.18980734050273895, | |
| "learning_rate": 6.93467336683417e-05, | |
| "loss": 0.5084, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.18042020499706268, | |
| "learning_rate": 6.834170854271357e-05, | |
| "loss": 0.5538, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 0.2088419497013092, | |
| "learning_rate": 6.733668341708544e-05, | |
| "loss": 0.4912, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.15689465403556824, | |
| "learning_rate": 6.633165829145729e-05, | |
| "loss": 0.5273, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.18145041167736053, | |
| "learning_rate": 6.532663316582915e-05, | |
| "loss": 0.5564, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.1750493198633194, | |
| "learning_rate": 6.4321608040201e-05, | |
| "loss": 0.4913, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 0.19720715284347534, | |
| "learning_rate": 6.331658291457287e-05, | |
| "loss": 0.5309, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.1915196031332016, | |
| "learning_rate": 6.231155778894473e-05, | |
| "loss": 0.5385, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 0.18675631284713745, | |
| "learning_rate": 6.130653266331658e-05, | |
| "loss": 0.5007, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.19217941164970398, | |
| "learning_rate": 6.030150753768844e-05, | |
| "loss": 0.4999, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 0.1979852020740509, | |
| "learning_rate": 5.929648241206031e-05, | |
| "loss": 0.4839, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 0.2420262098312378, | |
| "learning_rate": 5.829145728643216e-05, | |
| "loss": 0.4727, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 0.21145766973495483, | |
| "learning_rate": 5.728643216080403e-05, | |
| "loss": 0.4942, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.23028914630413055, | |
| "learning_rate": 5.628140703517588e-05, | |
| "loss": 0.4984, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 0.19866284728050232, | |
| "learning_rate": 5.527638190954774e-05, | |
| "loss": 0.4909, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 0.18788225948810577, | |
| "learning_rate": 5.4271356783919604e-05, | |
| "loss": 0.4889, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 0.20647984743118286, | |
| "learning_rate": 5.3266331658291455e-05, | |
| "loss": 0.4844, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.2238943725824356, | |
| "learning_rate": 5.226130653266332e-05, | |
| "loss": 0.4531, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 0.2138652801513672, | |
| "learning_rate": 5.125628140703518e-05, | |
| "loss": 0.4631, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.2267480194568634, | |
| "learning_rate": 5.0251256281407036e-05, | |
| "loss": 0.4617, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.069113651395789e+17, | |
| "train_batch_size": 10, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |