5e-5_4b_sft_nothink / trainer_state.json
Ibisbill's picture
Initial upload
8b0cbc6 verified
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.875912408759124,
"eval_steps": 1000,
"global_step": 51,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.058394160583941604,
"grad_norm": 4.656679619700339,
"learning_rate": 0.0,
"loss": 0.4427,
"step": 1
},
{
"epoch": 0.11678832116788321,
"grad_norm": 4.860123771865997,
"learning_rate": 8.333333333333334e-06,
"loss": 0.4479,
"step": 2
},
{
"epoch": 0.17518248175182483,
"grad_norm": 2.242956104917747,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.4042,
"step": 3
},
{
"epoch": 0.23357664233576642,
"grad_norm": 1.6608015519980013,
"learning_rate": 2.5e-05,
"loss": 0.3715,
"step": 4
},
{
"epoch": 0.291970802919708,
"grad_norm": 2.824986830456239,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.3487,
"step": 5
},
{
"epoch": 0.35036496350364965,
"grad_norm": 2.360090022024097,
"learning_rate": 4.166666666666667e-05,
"loss": 0.3541,
"step": 6
},
{
"epoch": 0.40875912408759124,
"grad_norm": 1.862994452379866,
"learning_rate": 5e-05,
"loss": 0.3371,
"step": 7
},
{
"epoch": 0.46715328467153283,
"grad_norm": 1.0596405353524974,
"learning_rate": 4.993910125649561e-05,
"loss": 0.328,
"step": 8
},
{
"epoch": 0.5255474452554745,
"grad_norm": 1.1579746499672514,
"learning_rate": 4.975670171853926e-05,
"loss": 0.3027,
"step": 9
},
{
"epoch": 0.583941605839416,
"grad_norm": 0.7003699851250199,
"learning_rate": 4.9453690018345144e-05,
"loss": 0.2944,
"step": 10
},
{
"epoch": 0.6423357664233577,
"grad_norm": 0.6949287751145983,
"learning_rate": 4.9031542398457974e-05,
"loss": 0.2966,
"step": 11
},
{
"epoch": 0.7007299270072993,
"grad_norm": 0.5584013509075192,
"learning_rate": 4.849231551964771e-05,
"loss": 0.29,
"step": 12
},
{
"epoch": 0.7591240875912408,
"grad_norm": 0.4653652334700578,
"learning_rate": 4.783863644106502e-05,
"loss": 0.2852,
"step": 13
},
{
"epoch": 0.8175182481751825,
"grad_norm": 0.4915332959658402,
"learning_rate": 4.707368982147318e-05,
"loss": 0.2793,
"step": 14
},
{
"epoch": 0.8759124087591241,
"grad_norm": 0.38234504853027035,
"learning_rate": 4.620120240391065e-05,
"loss": 0.2744,
"step": 15
},
{
"epoch": 0.9343065693430657,
"grad_norm": 0.4184062685451573,
"learning_rate": 4.522542485937369e-05,
"loss": 0.2768,
"step": 16
},
{
"epoch": 0.9927007299270073,
"grad_norm": 0.3704530279967013,
"learning_rate": 4.415111107797445e-05,
"loss": 0.2731,
"step": 17
},
{
"epoch": 1.0,
"grad_norm": 0.3704530279967013,
"learning_rate": 4.2983495008466276e-05,
"loss": 0.268,
"step": 18
},
{
"epoch": 1.0583941605839415,
"grad_norm": 0.7313410375782172,
"learning_rate": 4.172826515897146e-05,
"loss": 0.2522,
"step": 19
},
{
"epoch": 1.1167883211678833,
"grad_norm": 0.42450006771608867,
"learning_rate": 4.039153688314145e-05,
"loss": 0.2535,
"step": 20
},
{
"epoch": 1.1751824817518248,
"grad_norm": 0.35572497872199876,
"learning_rate": 3.897982258676867e-05,
"loss": 0.2468,
"step": 21
},
{
"epoch": 1.2335766423357664,
"grad_norm": 0.3840433540509707,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.2487,
"step": 22
},
{
"epoch": 1.2919708029197081,
"grad_norm": 0.3225052917728175,
"learning_rate": 3.5959278669726935e-05,
"loss": 0.2476,
"step": 23
},
{
"epoch": 1.3503649635036497,
"grad_norm": 0.3308473331591701,
"learning_rate": 3.436516483539781e-05,
"loss": 0.2437,
"step": 24
},
{
"epoch": 1.4087591240875912,
"grad_norm": 0.2817912935418844,
"learning_rate": 3.272542485937369e-05,
"loss": 0.2401,
"step": 25
},
{
"epoch": 1.4671532846715327,
"grad_norm": 0.3043640344168492,
"learning_rate": 3.104804738999169e-05,
"loss": 0.2443,
"step": 26
},
{
"epoch": 1.5255474452554745,
"grad_norm": 0.2334772712760174,
"learning_rate": 2.9341204441673266e-05,
"loss": 0.2428,
"step": 27
},
{
"epoch": 1.583941605839416,
"grad_norm": 0.252898158010675,
"learning_rate": 2.761321158169134e-05,
"loss": 0.2393,
"step": 28
},
{
"epoch": 1.6423357664233578,
"grad_norm": 0.19140920695753572,
"learning_rate": 2.587248741756253e-05,
"loss": 0.2341,
"step": 29
},
{
"epoch": 1.7007299270072993,
"grad_norm": 0.20503408790448174,
"learning_rate": 2.4127512582437485e-05,
"loss": 0.2364,
"step": 30
},
{
"epoch": 1.7591240875912408,
"grad_norm": 0.16892330641174735,
"learning_rate": 2.238678841830867e-05,
"loss": 0.2317,
"step": 31
},
{
"epoch": 1.8175182481751824,
"grad_norm": 0.18596878772651007,
"learning_rate": 2.0658795558326743e-05,
"loss": 0.2338,
"step": 32
},
{
"epoch": 1.8759124087591241,
"grad_norm": 0.19674830584353212,
"learning_rate": 1.895195261000831e-05,
"loss": 0.2385,
"step": 33
},
{
"epoch": 1.9343065693430657,
"grad_norm": 0.13767693529332986,
"learning_rate": 1.7274575140626318e-05,
"loss": 0.2254,
"step": 34
},
{
"epoch": 1.9927007299270074,
"grad_norm": 0.16469118823509277,
"learning_rate": 1.56348351646022e-05,
"loss": 0.2341,
"step": 35
},
{
"epoch": 2.0,
"grad_norm": 0.16469118823509277,
"learning_rate": 1.4040721330273062e-05,
"loss": 0.2609,
"step": 36
},
{
"epoch": 2.0583941605839415,
"grad_norm": 0.5065177528025752,
"learning_rate": 1.2500000000000006e-05,
"loss": 0.2222,
"step": 37
},
{
"epoch": 2.116788321167883,
"grad_norm": 0.14060766874048272,
"learning_rate": 1.1020177413231334e-05,
"loss": 0.2206,
"step": 38
},
{
"epoch": 2.1751824817518246,
"grad_norm": 0.1675154659569595,
"learning_rate": 9.608463116858542e-06,
"loss": 0.2257,
"step": 39
},
{
"epoch": 2.2335766423357666,
"grad_norm": 0.16579788203125537,
"learning_rate": 8.271734841028553e-06,
"loss": 0.2211,
"step": 40
},
{
"epoch": 2.291970802919708,
"grad_norm": 0.14053569348297154,
"learning_rate": 7.016504991533726e-06,
"loss": 0.2185,
"step": 41
},
{
"epoch": 2.3503649635036497,
"grad_norm": 0.1314158148612366,
"learning_rate": 5.848888922025553e-06,
"loss": 0.221,
"step": 42
},
{
"epoch": 2.408759124087591,
"grad_norm": 0.14777050841668893,
"learning_rate": 4.7745751406263165e-06,
"loss": 0.2219,
"step": 43
},
{
"epoch": 2.4671532846715327,
"grad_norm": 0.12484750949794649,
"learning_rate": 3.798797596089351e-06,
"loss": 0.2159,
"step": 44
},
{
"epoch": 2.5255474452554747,
"grad_norm": 0.11913135271242134,
"learning_rate": 2.9263101785268254e-06,
"loss": 0.2188,
"step": 45
},
{
"epoch": 2.5839416058394162,
"grad_norm": 0.11303754361927708,
"learning_rate": 2.1613635589349756e-06,
"loss": 0.222,
"step": 46
},
{
"epoch": 2.6423357664233578,
"grad_norm": 0.1181946037992105,
"learning_rate": 1.5076844803522922e-06,
"loss": 0.2218,
"step": 47
},
{
"epoch": 2.7007299270072993,
"grad_norm": 0.12180986717544033,
"learning_rate": 9.684576015420278e-07,
"loss": 0.2234,
"step": 48
},
{
"epoch": 2.759124087591241,
"grad_norm": 0.11708758326837156,
"learning_rate": 5.463099816548579e-07,
"loss": 0.2155,
"step": 49
},
{
"epoch": 2.8175182481751824,
"grad_norm": 0.11769771487209676,
"learning_rate": 2.4329828146074095e-07,
"loss": 0.2088,
"step": 50
},
{
"epoch": 2.875912408759124,
"grad_norm": 0.11487124367774278,
"learning_rate": 6.089874350439506e-08,
"loss": 0.2137,
"step": 51
}
],
"logging_steps": 1,
"max_steps": 51,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 20,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 121368125374464.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}