| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.603550295857988, | |
| "eval_steps": 5, | |
| "global_step": 110, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11834319526627218, | |
| "grad_norm": 0.11314600706100464, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.6482, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.11834319526627218, | |
| "eval_loss": 2.5026204586029053, | |
| "eval_runtime": 267.0476, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.154, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23668639053254437, | |
| "grad_norm": 0.12746678292751312, | |
| "learning_rate": 1.992981096013517e-05, | |
| "loss": 2.6479, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.23668639053254437, | |
| "eval_loss": 2.496476888656616, | |
| "eval_runtime": 267.3526, | |
| "eval_samples_per_second": 0.606, | |
| "eval_steps_per_second": 0.153, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.35502958579881655, | |
| "grad_norm": 0.15371470153331757, | |
| "learning_rate": 1.964635581908359e-05, | |
| "loss": 2.6197, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.35502958579881655, | |
| "eval_loss": 2.4890758991241455, | |
| "eval_runtime": 268.0217, | |
| "eval_samples_per_second": 0.604, | |
| "eval_steps_per_second": 0.153, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.47337278106508873, | |
| "grad_norm": 0.12974606454372406, | |
| "learning_rate": 1.9151456172430186e-05, | |
| "loss": 2.592, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.47337278106508873, | |
| "eval_loss": 2.4806125164031982, | |
| "eval_runtime": 267.5318, | |
| "eval_samples_per_second": 0.606, | |
| "eval_steps_per_second": 0.153, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.591715976331361, | |
| "grad_norm": 0.14419519901275635, | |
| "learning_rate": 1.845596003501826e-05, | |
| "loss": 2.6054, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.591715976331361, | |
| "eval_loss": 2.4711341857910156, | |
| "eval_runtime": 266.6209, | |
| "eval_samples_per_second": 0.608, | |
| "eval_steps_per_second": 0.154, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.7100591715976331, | |
| "grad_norm": 0.148014634847641, | |
| "learning_rate": 1.7575112421616203e-05, | |
| "loss": 2.5873, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7100591715976331, | |
| "eval_loss": 2.4612033367156982, | |
| "eval_runtime": 265.6543, | |
| "eval_samples_per_second": 0.61, | |
| "eval_steps_per_second": 0.154, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8284023668639053, | |
| "grad_norm": 0.1544329673051834, | |
| "learning_rate": 1.6528221181905217e-05, | |
| "loss": 2.6077, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.8284023668639053, | |
| "eval_loss": 2.4514548778533936, | |
| "eval_runtime": 267.0157, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.154, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.9467455621301775, | |
| "grad_norm": 0.14721575379371643, | |
| "learning_rate": 1.533823377964791e-05, | |
| "loss": 2.584, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.9467455621301775, | |
| "eval_loss": 2.442168951034546, | |
| "eval_runtime": 266.406, | |
| "eval_samples_per_second": 0.608, | |
| "eval_steps_per_second": 0.154, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0650887573964498, | |
| "grad_norm": 0.15120001137256622, | |
| "learning_rate": 1.4031234292879726e-05, | |
| "loss": 2.5831, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0650887573964498, | |
| "eval_loss": 2.433633327484131, | |
| "eval_runtime": 266.429, | |
| "eval_samples_per_second": 0.608, | |
| "eval_steps_per_second": 0.154, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.183431952662722, | |
| "grad_norm": 0.1626676619052887, | |
| "learning_rate": 1.2635871660690677e-05, | |
| "loss": 2.5507, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.183431952662722, | |
| "eval_loss": 2.425856828689575, | |
| "eval_runtime": 266.5163, | |
| "eval_samples_per_second": 0.608, | |
| "eval_steps_per_second": 0.154, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.301775147928994, | |
| "grad_norm": 0.15653495490550995, | |
| "learning_rate": 1.1182731709213658e-05, | |
| "loss": 2.5446, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.301775147928994, | |
| "eval_loss": 2.4187350273132324, | |
| "eval_runtime": 266.7761, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.154, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.4201183431952662, | |
| "grad_norm": 0.18180669844150543, | |
| "learning_rate": 9.703666721774403e-06, | |
| "loss": 2.5259, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.4201183431952662, | |
| "eval_loss": 2.412436008453369, | |
| "eval_runtime": 267.1031, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.153, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.1720452457666397, | |
| "learning_rate": 8.231097248774273e-06, | |
| "loss": 2.5576, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "eval_loss": 2.407033920288086, | |
| "eval_runtime": 266.8119, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.154, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.6568047337278107, | |
| "grad_norm": 0.1620107889175415, | |
| "learning_rate": 6.797301461371626e-06, | |
| "loss": 2.5783, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.6568047337278107, | |
| "eval_loss": 2.402543783187866, | |
| "eval_runtime": 267.0918, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.154, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.7751479289940828, | |
| "grad_norm": 0.1715419441461563, | |
| "learning_rate": 5.43370762606287e-06, | |
| "loss": 2.5542, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.7751479289940828, | |
| "eval_loss": 2.398886203765869, | |
| "eval_runtime": 267.0041, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.154, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.893491124260355, | |
| "grad_norm": 0.16874942183494568, | |
| "learning_rate": 4.170205208855281e-06, | |
| "loss": 2.5091, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.893491124260355, | |
| "eval_loss": 2.3959767818450928, | |
| "eval_runtime": 266.8823, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.154, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0118343195266273, | |
| "grad_norm": 0.16341295838356018, | |
| "learning_rate": 3.0344897093700333e-06, | |
| "loss": 2.528, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.0118343195266273, | |
| "eval_loss": 2.393803596496582, | |
| "eval_runtime": 266.9041, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.154, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.1301775147928996, | |
| "grad_norm": 0.16567575931549072, | |
| "learning_rate": 2.0514555858664663e-06, | |
| "loss": 2.5251, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.1301775147928996, | |
| "eval_loss": 2.3922572135925293, | |
| "eval_runtime": 267.9022, | |
| "eval_samples_per_second": 0.605, | |
| "eval_steps_per_second": 0.153, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.2485207100591715, | |
| "grad_norm": 0.16893596947193146, | |
| "learning_rate": 1.2426505780436326e-06, | |
| "loss": 2.5235, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.2485207100591715, | |
| "eval_loss": 2.3912644386291504, | |
| "eval_runtime": 266.3428, | |
| "eval_samples_per_second": 0.608, | |
| "eval_steps_per_second": 0.154, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.366863905325444, | |
| "grad_norm": 0.16508392989635468, | |
| "learning_rate": 6.258033886587911e-07, | |
| "loss": 2.5154, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.366863905325444, | |
| "eval_loss": 2.3907008171081543, | |
| "eval_runtime": 266.5099, | |
| "eval_samples_per_second": 0.608, | |
| "eval_steps_per_second": 0.154, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.485207100591716, | |
| "grad_norm": 0.15974663197994232, | |
| "learning_rate": 2.1443507700495968e-07, | |
| "loss": 2.5333, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.485207100591716, | |
| "eval_loss": 2.3904457092285156, | |
| "eval_runtime": 265.5021, | |
| "eval_samples_per_second": 0.61, | |
| "eval_steps_per_second": 0.154, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.603550295857988, | |
| "grad_norm": 0.17541921138763428, | |
| "learning_rate": 1.7562682356786488e-08, | |
| "loss": 2.5098, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.603550295857988, | |
| "eval_loss": 2.390380382537842, | |
| "eval_runtime": 266.7873, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.154, | |
| "step": 110 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 112, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 5, | |
| "total_flos": 5.736198700007424e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |