| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.4624277456647399, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004624277456647399, | |
| "eval_loss": 4.693221569061279, | |
| "eval_runtime": 23.3821, | |
| "eval_samples_per_second": 15.61, | |
| "eval_steps_per_second": 1.967, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.013872832369942197, | |
| "grad_norm": 6.138703346252441, | |
| "learning_rate": 1.5e-05, | |
| "loss": 18.4969, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.027745664739884393, | |
| "grad_norm": 6.098344802856445, | |
| "learning_rate": 3e-05, | |
| "loss": 18.5023, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.04161849710982659, | |
| "grad_norm": 8.08814525604248, | |
| "learning_rate": 4.5e-05, | |
| "loss": 18.2983, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04161849710982659, | |
| "eval_loss": 4.422519207000732, | |
| "eval_runtime": 23.2409, | |
| "eval_samples_per_second": 15.705, | |
| "eval_steps_per_second": 1.979, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.055491329479768786, | |
| "grad_norm": 13.294960975646973, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 17.6258, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.06936416184971098, | |
| "grad_norm": 12.818150520324707, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 13.5799, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08323699421965318, | |
| "grad_norm": 11.56785774230957, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 11.4697, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08323699421965318, | |
| "eval_loss": 2.661552906036377, | |
| "eval_runtime": 23.3554, | |
| "eval_samples_per_second": 15.628, | |
| "eval_steps_per_second": 1.97, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.09710982658959537, | |
| "grad_norm": 11.117833137512207, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 9.6316, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.11098265895953757, | |
| "grad_norm": 14.03640079498291, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 10.2925, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.12485549132947976, | |
| "grad_norm": 12.930510520935059, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 9.8234, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.12485549132947976, | |
| "eval_loss": 2.4066836833953857, | |
| "eval_runtime": 23.3713, | |
| "eval_samples_per_second": 15.617, | |
| "eval_steps_per_second": 1.968, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.13872832369942195, | |
| "grad_norm": 13.057132720947266, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 8.9095, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15260115606936417, | |
| "grad_norm": 13.501303672790527, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 9.0641, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.16647398843930636, | |
| "grad_norm": 13.990789413452148, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 8.5, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.16647398843930636, | |
| "eval_loss": 2.281359910964966, | |
| "eval_runtime": 23.3762, | |
| "eval_samples_per_second": 15.614, | |
| "eval_steps_per_second": 1.968, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.18034682080924855, | |
| "grad_norm": 15.291366577148438, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 9.3509, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.19421965317919074, | |
| "grad_norm": 12.936572074890137, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 9.5222, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.20809248554913296, | |
| "grad_norm": 12.182962417602539, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 9.2995, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.20809248554913296, | |
| "eval_loss": 2.2252280712127686, | |
| "eval_runtime": 23.3704, | |
| "eval_samples_per_second": 15.618, | |
| "eval_steps_per_second": 1.968, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.22196531791907514, | |
| "grad_norm": 13.763567924499512, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 8.9847, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.23583815028901733, | |
| "grad_norm": 10.402312278747559, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 8.2846, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.24971098265895952, | |
| "grad_norm": 12.686690330505371, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 9.2072, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.24971098265895952, | |
| "eval_loss": 2.1948208808898926, | |
| "eval_runtime": 23.3918, | |
| "eval_samples_per_second": 15.604, | |
| "eval_steps_per_second": 1.967, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.2635838150289017, | |
| "grad_norm": 11.070029258728027, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 8.5202, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.2774566473988439, | |
| "grad_norm": 14.184294700622559, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 9.3465, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.29132947976878615, | |
| "grad_norm": 11.007719993591309, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 9.118, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.29132947976878615, | |
| "eval_loss": 2.174855947494507, | |
| "eval_runtime": 23.3884, | |
| "eval_samples_per_second": 15.606, | |
| "eval_steps_per_second": 1.967, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.30520231213872834, | |
| "grad_norm": 13.206997871398926, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 8.8766, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.3190751445086705, | |
| "grad_norm": 10.500737190246582, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 7.9143, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.3329479768786127, | |
| "grad_norm": 11.877433776855469, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 8.8372, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3329479768786127, | |
| "eval_loss": 2.1602916717529297, | |
| "eval_runtime": 23.3824, | |
| "eval_samples_per_second": 15.61, | |
| "eval_steps_per_second": 1.967, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3468208092485549, | |
| "grad_norm": 10.657315254211426, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 8.5161, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.3606936416184971, | |
| "grad_norm": 11.294054985046387, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 8.1042, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.3745664739884393, | |
| "grad_norm": 12.412280082702637, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 8.4474, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.3745664739884393, | |
| "eval_loss": 2.152019739151001, | |
| "eval_runtime": 23.3691, | |
| "eval_samples_per_second": 15.619, | |
| "eval_steps_per_second": 1.968, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.3884393063583815, | |
| "grad_norm": 11.942791938781738, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 9.0616, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.4023121387283237, | |
| "grad_norm": 11.127038955688477, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 8.395, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.4161849710982659, | |
| "grad_norm": 13.24832534790039, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 9.1845, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4161849710982659, | |
| "eval_loss": 2.148911237716675, | |
| "eval_runtime": 23.3723, | |
| "eval_samples_per_second": 15.617, | |
| "eval_steps_per_second": 1.968, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4300578034682081, | |
| "grad_norm": 11.400464057922363, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 8.8267, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.4439306358381503, | |
| "grad_norm": 13.761180877685547, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 8.0885, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.4578034682080925, | |
| "grad_norm": 11.87588882446289, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 9.3402, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.4578034682080925, | |
| "eval_loss": 2.1476869583129883, | |
| "eval_runtime": 23.3804, | |
| "eval_samples_per_second": 15.611, | |
| "eval_steps_per_second": 1.967, | |
| "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": 1.309190765150208e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |