| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.009048545446319504, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.048545446319505e-05, | |
| "eval_loss": 2.0614755153656006, | |
| "eval_runtime": 1431.9492, | |
| "eval_samples_per_second": 12.998, | |
| "eval_steps_per_second": 1.625, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00027145636338958513, | |
| "grad_norm": 3.0326344966888428, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.1455, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0005429127267791703, | |
| "grad_norm": 2.6346302032470703, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2541, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0008143690901687554, | |
| "grad_norm": 2.24157977104187, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.1746, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0008143690901687554, | |
| "eval_loss": 1.376136302947998, | |
| "eval_runtime": 1439.157, | |
| "eval_samples_per_second": 12.933, | |
| "eval_steps_per_second": 1.617, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0010858254535583405, | |
| "grad_norm": 2.4520862102508545, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.9936, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0013572818169479257, | |
| "grad_norm": 2.1917343139648438, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.7332, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0016287381803375108, | |
| "grad_norm": 2.0286247730255127, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.6707, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0016287381803375108, | |
| "eval_loss": 0.708565354347229, | |
| "eval_runtime": 1438.9184, | |
| "eval_samples_per_second": 12.935, | |
| "eval_steps_per_second": 1.617, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.001900194543727096, | |
| "grad_norm": 2.2299928665161133, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.5779, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.002171650907116681, | |
| "grad_norm": 3.2380049228668213, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.6919, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.002443107270506266, | |
| "grad_norm": 2.030574321746826, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.4923, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.002443107270506266, | |
| "eval_loss": 0.5967397689819336, | |
| "eval_runtime": 1439.4951, | |
| "eval_samples_per_second": 12.93, | |
| "eval_steps_per_second": 1.617, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0027145636338958513, | |
| "grad_norm": 3.188636541366577, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.4312, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0029860199972854364, | |
| "grad_norm": 2.3065080642700195, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.6207, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0032574763606750216, | |
| "grad_norm": 1.9325222969055176, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.5588, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0032574763606750216, | |
| "eval_loss": 0.5455918908119202, | |
| "eval_runtime": 1439.2701, | |
| "eval_samples_per_second": 12.932, | |
| "eval_steps_per_second": 1.617, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0035289327240646067, | |
| "grad_norm": 2.0116512775421143, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.5004, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.003800389087454192, | |
| "grad_norm": 2.0821337699890137, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.4352, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0040718454508437765, | |
| "grad_norm": 1.7970950603485107, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.533, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0040718454508437765, | |
| "eval_loss": 0.5220184326171875, | |
| "eval_runtime": 1438.7352, | |
| "eval_samples_per_second": 12.937, | |
| "eval_steps_per_second": 1.617, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004343301814233362, | |
| "grad_norm": 2.69805645942688, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.4621, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.004614758177622947, | |
| "grad_norm": 1.981372594833374, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.6349, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.004886214541012532, | |
| "grad_norm": 2.258195161819458, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.5486, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.004886214541012532, | |
| "eval_loss": 0.5081658363342285, | |
| "eval_runtime": 1439.3002, | |
| "eval_samples_per_second": 12.932, | |
| "eval_steps_per_second": 1.617, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.005157670904402117, | |
| "grad_norm": 1.517822504043579, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.5922, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.005429127267791703, | |
| "grad_norm": 1.7860591411590576, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.437, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.005700583631181287, | |
| "grad_norm": 4.32674503326416, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.5061, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.005700583631181287, | |
| "eval_loss": 0.4976564645767212, | |
| "eval_runtime": 1439.5584, | |
| "eval_samples_per_second": 12.93, | |
| "eval_steps_per_second": 1.616, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.005972039994570873, | |
| "grad_norm": 1.9437718391418457, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.5445, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.006243496357960458, | |
| "grad_norm": 1.3337701559066772, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.4558, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.006514952721350043, | |
| "grad_norm": 2.283390998840332, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.5451, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.006514952721350043, | |
| "eval_loss": 0.4924093782901764, | |
| "eval_runtime": 1439.0551, | |
| "eval_samples_per_second": 12.934, | |
| "eval_steps_per_second": 1.617, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.006786409084739628, | |
| "grad_norm": 2.103365421295166, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.5615, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.007057865448129213, | |
| "grad_norm": 1.8247758150100708, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.4278, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.007329321811518798, | |
| "grad_norm": 2.0524744987487793, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.5058, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.007329321811518798, | |
| "eval_loss": 0.48786625266075134, | |
| "eval_runtime": 1439.1036, | |
| "eval_samples_per_second": 12.934, | |
| "eval_steps_per_second": 1.617, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.007600778174908384, | |
| "grad_norm": 2.635622501373291, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.5012, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.00787223453829797, | |
| "grad_norm": 1.4843945503234863, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.4902, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.008143690901687553, | |
| "grad_norm": 1.9264837503433228, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.4289, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.008143690901687553, | |
| "eval_loss": 0.48586252331733704, | |
| "eval_runtime": 1438.3431, | |
| "eval_samples_per_second": 12.941, | |
| "eval_steps_per_second": 1.618, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.008415147265077139, | |
| "grad_norm": 2.084927797317505, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.5326, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.008686603628466724, | |
| "grad_norm": 1.7900018692016602, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.4558, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.00895805999185631, | |
| "grad_norm": 1.933836817741394, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.5505, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.00895805999185631, | |
| "eval_loss": 0.4853900074958801, | |
| "eval_runtime": 1438.1766, | |
| "eval_samples_per_second": 12.942, | |
| "eval_steps_per_second": 1.618, | |
| "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": 1.44418812198912e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |