| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5805515239477503, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005805515239477504, | |
| "eval_loss": 1.2620487213134766, | |
| "eval_runtime": 8.1244, | |
| "eval_samples_per_second": 35.695, | |
| "eval_steps_per_second": 4.554, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01741654571843251, | |
| "grad_norm": 0.4328065812587738, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.2349, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.03483309143686502, | |
| "grad_norm": 0.46443697810173035, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2858, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05224963715529753, | |
| "grad_norm": 0.44606202840805054, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.2235, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05224963715529753, | |
| "eval_loss": 1.2377574443817139, | |
| "eval_runtime": 8.1496, | |
| "eval_samples_per_second": 35.584, | |
| "eval_steps_per_second": 4.54, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.06966618287373004, | |
| "grad_norm": 0.5522386431694031, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.2714, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08708272859216255, | |
| "grad_norm": 0.5637324452400208, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 1.1623, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.10449927431059507, | |
| "grad_norm": 0.5763392448425293, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.0162, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.10449927431059507, | |
| "eval_loss": 1.0046803951263428, | |
| "eval_runtime": 8.16, | |
| "eval_samples_per_second": 35.539, | |
| "eval_steps_per_second": 4.534, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.12191582002902758, | |
| "grad_norm": 0.5673230886459351, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.9585, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.13933236574746008, | |
| "grad_norm": 0.47465190291404724, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.8755, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1567489114658926, | |
| "grad_norm": 0.5033220052719116, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.788, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1567489114658926, | |
| "eval_loss": 0.743035078048706, | |
| "eval_runtime": 8.1945, | |
| "eval_samples_per_second": 35.389, | |
| "eval_steps_per_second": 4.515, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1741654571843251, | |
| "grad_norm": 0.4010712802410126, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.6837, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19158200290275762, | |
| "grad_norm": 0.38208135962486267, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.6335, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.20899854862119013, | |
| "grad_norm": 0.3118175268173218, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.5961, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.20899854862119013, | |
| "eval_loss": 0.5937825441360474, | |
| "eval_runtime": 8.2105, | |
| "eval_samples_per_second": 35.32, | |
| "eval_steps_per_second": 4.506, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.22641509433962265, | |
| "grad_norm": 0.26881372928619385, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.61, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.24383164005805516, | |
| "grad_norm": 0.2959866225719452, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.5413, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.2612481857764877, | |
| "grad_norm": 0.31090644001960754, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.5359, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2612481857764877, | |
| "eval_loss": 0.5296854972839355, | |
| "eval_runtime": 8.2316, | |
| "eval_samples_per_second": 35.23, | |
| "eval_steps_per_second": 4.495, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.27866473149492016, | |
| "grad_norm": 0.30757585167884827, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.4991, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.2960812772133527, | |
| "grad_norm": 0.2986898720264435, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.5406, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.3134978229317852, | |
| "grad_norm": 0.29928526282310486, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.5209, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.3134978229317852, | |
| "eval_loss": 0.499483197927475, | |
| "eval_runtime": 8.228, | |
| "eval_samples_per_second": 35.245, | |
| "eval_steps_per_second": 4.497, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.3309143686502177, | |
| "grad_norm": 0.3250415325164795, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.5339, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.3483309143686502, | |
| "grad_norm": 0.2665368616580963, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.4771, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.36574746008708275, | |
| "grad_norm": 0.2685997784137726, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.4493, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.36574746008708275, | |
| "eval_loss": 0.48386040329933167, | |
| "eval_runtime": 8.2199, | |
| "eval_samples_per_second": 35.28, | |
| "eval_steps_per_second": 4.501, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.38316400580551524, | |
| "grad_norm": 0.3184368908405304, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.4958, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.4005805515239477, | |
| "grad_norm": 0.2603450417518616, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.4831, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.41799709724238027, | |
| "grad_norm": 0.3131684362888336, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.4193, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.41799709724238027, | |
| "eval_loss": 0.4739016890525818, | |
| "eval_runtime": 8.222, | |
| "eval_samples_per_second": 35.271, | |
| "eval_steps_per_second": 4.5, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.43541364296081275, | |
| "grad_norm": 0.2975787818431854, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.4181, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4528301886792453, | |
| "grad_norm": 0.3197750747203827, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.4859, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.4702467343976778, | |
| "grad_norm": 0.29346856474876404, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.4102, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.4702467343976778, | |
| "eval_loss": 0.46982458233833313, | |
| "eval_runtime": 8.2212, | |
| "eval_samples_per_second": 35.274, | |
| "eval_steps_per_second": 4.501, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.4876632801161103, | |
| "grad_norm": 0.30700182914733887, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.5111, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.5050798258345428, | |
| "grad_norm": 0.3279879689216614, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.4594, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.5224963715529753, | |
| "grad_norm": 0.2903614938259125, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.4378, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5224963715529753, | |
| "eval_loss": 0.4680546224117279, | |
| "eval_runtime": 8.2219, | |
| "eval_samples_per_second": 35.272, | |
| "eval_steps_per_second": 4.5, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5399129172714079, | |
| "grad_norm": 0.32795748114585876, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.503, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.5573294629898403, | |
| "grad_norm": 0.3038257658481598, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.4698, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.5747460087082729, | |
| "grad_norm": 0.30126094818115234, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.4015, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.5747460087082729, | |
| "eval_loss": 0.4675796627998352, | |
| "eval_runtime": 8.2149, | |
| "eval_samples_per_second": 35.302, | |
| "eval_steps_per_second": 4.504, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.20235311726592e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |