| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2302325581395349, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002325581395348837, | |
| "eval_loss": 5.892056941986084, | |
| "eval_runtime": 19.7546, | |
| "eval_samples_per_second": 36.65, | |
| "eval_steps_per_second": 4.607, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0069767441860465115, | |
| "grad_norm": 3.488095760345459, | |
| "learning_rate": 1.5e-05, | |
| "loss": 5.9292, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.013953488372093023, | |
| "grad_norm": 2.8731729984283447, | |
| "learning_rate": 3e-05, | |
| "loss": 5.7853, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.020930232558139535, | |
| "grad_norm": 4.2616777420043945, | |
| "learning_rate": 4.5e-05, | |
| "loss": 5.9753, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.020930232558139535, | |
| "eval_loss": 5.681079387664795, | |
| "eval_runtime": 19.8386, | |
| "eval_samples_per_second": 36.494, | |
| "eval_steps_per_second": 4.587, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.027906976744186046, | |
| "grad_norm": 4.38749361038208, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 5.632, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03488372093023256, | |
| "grad_norm": 4.592843055725098, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 4.9077, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04186046511627907, | |
| "grad_norm": 4.174410820007324, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 4.5093, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04186046511627907, | |
| "eval_loss": 4.481910705566406, | |
| "eval_runtime": 19.8608, | |
| "eval_samples_per_second": 36.454, | |
| "eval_steps_per_second": 4.582, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04883720930232558, | |
| "grad_norm": 4.76424503326416, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 4.4965, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05581395348837209, | |
| "grad_norm": 4.088315963745117, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 3.9977, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06279069767441861, | |
| "grad_norm": 3.9060111045837402, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 3.9268, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06279069767441861, | |
| "eval_loss": 3.678138256072998, | |
| "eval_runtime": 19.8735, | |
| "eval_samples_per_second": 36.43, | |
| "eval_steps_per_second": 4.579, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06976744186046512, | |
| "grad_norm": 4.486081123352051, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 3.6693, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07674418604651163, | |
| "grad_norm": 3.4287188053131104, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 3.0881, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.08372093023255814, | |
| "grad_norm": 3.1772103309631348, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 3.2421, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.08372093023255814, | |
| "eval_loss": 3.141343832015991, | |
| "eval_runtime": 19.8825, | |
| "eval_samples_per_second": 36.414, | |
| "eval_steps_per_second": 4.577, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09069767441860466, | |
| "grad_norm": 3.955404758453369, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 3.0893, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.09767441860465116, | |
| "grad_norm": 3.391456365585327, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 2.924, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.10465116279069768, | |
| "grad_norm": 3.183100461959839, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 2.8891, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10465116279069768, | |
| "eval_loss": 2.8821494579315186, | |
| "eval_runtime": 19.8866, | |
| "eval_samples_per_second": 36.406, | |
| "eval_steps_per_second": 4.576, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11162790697674418, | |
| "grad_norm": 3.60516095161438, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 2.7192, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1186046511627907, | |
| "grad_norm": 3.239312171936035, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 2.7269, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.12558139534883722, | |
| "grad_norm": 2.7312939167022705, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 2.6379, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.12558139534883722, | |
| "eval_loss": 2.7178566455841064, | |
| "eval_runtime": 19.8989, | |
| "eval_samples_per_second": 36.384, | |
| "eval_steps_per_second": 4.573, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.1325581395348837, | |
| "grad_norm": 3.0902271270751953, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 2.8834, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.13953488372093023, | |
| "grad_norm": 2.8137857913970947, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 2.5538, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.14651162790697675, | |
| "grad_norm": 2.7316954135894775, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 2.5681, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.14651162790697675, | |
| "eval_loss": 2.624565601348877, | |
| "eval_runtime": 19.8773, | |
| "eval_samples_per_second": 36.423, | |
| "eval_steps_per_second": 4.578, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.15348837209302327, | |
| "grad_norm": 2.953460693359375, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 2.3964, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.16046511627906976, | |
| "grad_norm": 2.805899143218994, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 2.4172, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.16744186046511628, | |
| "grad_norm": 3.291818857192993, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 2.522, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.16744186046511628, | |
| "eval_loss": 2.5648176670074463, | |
| "eval_runtime": 19.8785, | |
| "eval_samples_per_second": 36.421, | |
| "eval_steps_per_second": 4.578, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1744186046511628, | |
| "grad_norm": 2.5910260677337646, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 2.7511, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1813953488372093, | |
| "grad_norm": 2.6901168823242188, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 2.6676, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.1883720930232558, | |
| "grad_norm": 2.912105083465576, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 2.5736, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.1883720930232558, | |
| "eval_loss": 2.537811517715454, | |
| "eval_runtime": 19.8766, | |
| "eval_samples_per_second": 36.425, | |
| "eval_steps_per_second": 4.578, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.19534883720930232, | |
| "grad_norm": 2.613006830215454, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 2.3515, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.20232558139534884, | |
| "grad_norm": 3.0070550441741943, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 2.6264, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.20930232558139536, | |
| "grad_norm": 2.684951066970825, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 2.4758, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.20930232558139536, | |
| "eval_loss": 2.527015447616577, | |
| "eval_runtime": 19.8758, | |
| "eval_samples_per_second": 36.426, | |
| "eval_steps_per_second": 4.578, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.21627906976744185, | |
| "grad_norm": 3.2055976390838623, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 2.673, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.22325581395348837, | |
| "grad_norm": 2.951667308807373, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 2.5266, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2302325581395349, | |
| "grad_norm": 2.4879679679870605, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 2.4875, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.2302325581395349, | |
| "eval_loss": 2.5238454341888428, | |
| "eval_runtime": 19.9264, | |
| "eval_samples_per_second": 36.334, | |
| "eval_steps_per_second": 4.567, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.586412296909619e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |