lora-imdb_reviews / trainer_state.json
igzi's picture
Upload latest checkpoint for imdb_reviews
4ab04af verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.10080645161290322,
"eval_steps": 500,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0010080645161290322,
"grad_norm": 1.5859375,
"learning_rate": 0.00019800000000000002,
"loss": 1.1003,
"step": 50
},
{
"epoch": 0.0020161290322580645,
"grad_norm": 0.0,
"learning_rate": 0.000196,
"loss": 1.2089,
"step": 100
},
{
"epoch": 0.0030241935483870967,
"grad_norm": 0.00189208984375,
"learning_rate": 0.000194,
"loss": 1.3291,
"step": 150
},
{
"epoch": 0.004032258064516129,
"grad_norm": 0.028564453125,
"learning_rate": 0.000192,
"loss": 0.87,
"step": 200
},
{
"epoch": 0.005040322580645161,
"grad_norm": 1.6484375,
"learning_rate": 0.00019,
"loss": 1.0644,
"step": 250
},
{
"epoch": 0.006048387096774193,
"grad_norm": 0.00119781494140625,
"learning_rate": 0.000188,
"loss": 0.9578,
"step": 300
},
{
"epoch": 0.007056451612903226,
"grad_norm": 0.8359375,
"learning_rate": 0.00018600000000000002,
"loss": 0.7294,
"step": 350
},
{
"epoch": 0.008064516129032258,
"grad_norm": 1.78125,
"learning_rate": 0.00018400000000000003,
"loss": 1.126,
"step": 400
},
{
"epoch": 0.009072580645161291,
"grad_norm": 0.0113525390625,
"learning_rate": 0.000182,
"loss": 1.659,
"step": 450
},
{
"epoch": 0.010080645161290322,
"grad_norm": 0.00714111328125,
"learning_rate": 0.00018,
"loss": 1.2084,
"step": 500
},
{
"epoch": 0.011088709677419355,
"grad_norm": 0.055908203125,
"learning_rate": 0.00017800000000000002,
"loss": 0.9386,
"step": 550
},
{
"epoch": 0.012096774193548387,
"grad_norm": 0.828125,
"learning_rate": 0.00017600000000000002,
"loss": 1.4331,
"step": 600
},
{
"epoch": 0.01310483870967742,
"grad_norm": 15.4375,
"learning_rate": 0.000174,
"loss": 1.1943,
"step": 650
},
{
"epoch": 0.014112903225806451,
"grad_norm": 0.0,
"learning_rate": 0.000172,
"loss": 0.7673,
"step": 700
},
{
"epoch": 0.015120967741935484,
"grad_norm": 0.0096435546875,
"learning_rate": 0.00017,
"loss": 0.9278,
"step": 750
},
{
"epoch": 0.016129032258064516,
"grad_norm": 0.00262451171875,
"learning_rate": 0.000168,
"loss": 0.8818,
"step": 800
},
{
"epoch": 0.017137096774193547,
"grad_norm": 0.0098876953125,
"learning_rate": 0.000166,
"loss": 1.0701,
"step": 850
},
{
"epoch": 0.018145161290322582,
"grad_norm": 0.0025177001953125,
"learning_rate": 0.000164,
"loss": 1.0132,
"step": 900
},
{
"epoch": 0.019153225806451613,
"grad_norm": 0.0162353515625,
"learning_rate": 0.000162,
"loss": 0.909,
"step": 950
},
{
"epoch": 0.020161290322580645,
"grad_norm": 0.0,
"learning_rate": 0.00016,
"loss": 1.1242,
"step": 1000
},
{
"epoch": 0.021169354838709676,
"grad_norm": 0.0,
"learning_rate": 0.00015800000000000002,
"loss": 1.4603,
"step": 1050
},
{
"epoch": 0.02217741935483871,
"grad_norm": 0.66796875,
"learning_rate": 0.00015600000000000002,
"loss": 0.7965,
"step": 1100
},
{
"epoch": 0.023185483870967742,
"grad_norm": 0.00054168701171875,
"learning_rate": 0.000154,
"loss": 0.7828,
"step": 1150
},
{
"epoch": 0.024193548387096774,
"grad_norm": 0.9921875,
"learning_rate": 0.000152,
"loss": 0.9853,
"step": 1200
},
{
"epoch": 0.025201612903225805,
"grad_norm": 0.00689697265625,
"learning_rate": 0.00015000000000000001,
"loss": 1.375,
"step": 1250
},
{
"epoch": 0.02620967741935484,
"grad_norm": 0.0,
"learning_rate": 0.000148,
"loss": 0.9719,
"step": 1300
},
{
"epoch": 0.02721774193548387,
"grad_norm": 0.0,
"learning_rate": 0.000146,
"loss": 1.2151,
"step": 1350
},
{
"epoch": 0.028225806451612902,
"grad_norm": 0.00982666015625,
"learning_rate": 0.000144,
"loss": 0.7872,
"step": 1400
},
{
"epoch": 0.029233870967741934,
"grad_norm": 0.00653076171875,
"learning_rate": 0.000142,
"loss": 1.0166,
"step": 1450
},
{
"epoch": 0.03024193548387097,
"grad_norm": 0.00665283203125,
"learning_rate": 0.00014,
"loss": 1.0774,
"step": 1500
},
{
"epoch": 0.03125,
"grad_norm": 0.00933837890625,
"learning_rate": 0.000138,
"loss": 1.0596,
"step": 1550
},
{
"epoch": 0.03225806451612903,
"grad_norm": 0.85546875,
"learning_rate": 0.00013600000000000003,
"loss": 1.048,
"step": 1600
},
{
"epoch": 0.03326612903225806,
"grad_norm": 0.0,
"learning_rate": 0.000134,
"loss": 0.8372,
"step": 1650
},
{
"epoch": 0.034274193548387094,
"grad_norm": 0.81640625,
"learning_rate": 0.000132,
"loss": 1.4669,
"step": 1700
},
{
"epoch": 0.03528225806451613,
"grad_norm": 0.0,
"learning_rate": 0.00013000000000000002,
"loss": 0.7405,
"step": 1750
},
{
"epoch": 0.036290322580645164,
"grad_norm": 0.0,
"learning_rate": 0.00012800000000000002,
"loss": 0.988,
"step": 1800
},
{
"epoch": 0.037298387096774195,
"grad_norm": 0.0096435546875,
"learning_rate": 0.000126,
"loss": 1.1337,
"step": 1850
},
{
"epoch": 0.038306451612903226,
"grad_norm": 0.006011962890625,
"learning_rate": 0.000124,
"loss": 0.8881,
"step": 1900
},
{
"epoch": 0.03931451612903226,
"grad_norm": 0.02197265625,
"learning_rate": 0.000122,
"loss": 0.9004,
"step": 1950
},
{
"epoch": 0.04032258064516129,
"grad_norm": 0.002838134765625,
"learning_rate": 0.00012,
"loss": 0.9567,
"step": 2000
},
{
"epoch": 0.04133064516129032,
"grad_norm": 0.0230712890625,
"learning_rate": 0.000118,
"loss": 0.9933,
"step": 2050
},
{
"epoch": 0.04233870967741935,
"grad_norm": 0.002471923828125,
"learning_rate": 0.000116,
"loss": 1.0882,
"step": 2100
},
{
"epoch": 0.04334677419354839,
"grad_norm": 0.01080322265625,
"learning_rate": 0.00011399999999999999,
"loss": 0.9519,
"step": 2150
},
{
"epoch": 0.04435483870967742,
"grad_norm": 0.032958984375,
"learning_rate": 0.00011200000000000001,
"loss": 0.7841,
"step": 2200
},
{
"epoch": 0.04536290322580645,
"grad_norm": 0.00653076171875,
"learning_rate": 0.00011000000000000002,
"loss": 0.7603,
"step": 2250
},
{
"epoch": 0.046370967741935484,
"grad_norm": 0.012939453125,
"learning_rate": 0.00010800000000000001,
"loss": 1.2195,
"step": 2300
},
{
"epoch": 0.047379032258064516,
"grad_norm": 0.00433349609375,
"learning_rate": 0.00010600000000000002,
"loss": 1.2472,
"step": 2350
},
{
"epoch": 0.04838709677419355,
"grad_norm": 0.04052734375,
"learning_rate": 0.00010400000000000001,
"loss": 0.9298,
"step": 2400
},
{
"epoch": 0.04939516129032258,
"grad_norm": 0.0,
"learning_rate": 0.00010200000000000001,
"loss": 1.3759,
"step": 2450
},
{
"epoch": 0.05040322580645161,
"grad_norm": 0.00701904296875,
"learning_rate": 0.0001,
"loss": 0.7772,
"step": 2500
},
{
"epoch": 0.05141129032258065,
"grad_norm": 0.0120849609375,
"learning_rate": 9.8e-05,
"loss": 1.1628,
"step": 2550
},
{
"epoch": 0.05241935483870968,
"grad_norm": 0.041748046875,
"learning_rate": 9.6e-05,
"loss": 1.064,
"step": 2600
},
{
"epoch": 0.05342741935483871,
"grad_norm": 0.0,
"learning_rate": 9.4e-05,
"loss": 0.9408,
"step": 2650
},
{
"epoch": 0.05443548387096774,
"grad_norm": 1.0078125,
"learning_rate": 9.200000000000001e-05,
"loss": 0.933,
"step": 2700
},
{
"epoch": 0.055443548387096774,
"grad_norm": 0.03466796875,
"learning_rate": 9e-05,
"loss": 0.918,
"step": 2750
},
{
"epoch": 0.056451612903225805,
"grad_norm": 0.0322265625,
"learning_rate": 8.800000000000001e-05,
"loss": 0.8227,
"step": 2800
},
{
"epoch": 0.057459677419354836,
"grad_norm": 0.84765625,
"learning_rate": 8.6e-05,
"loss": 0.974,
"step": 2850
},
{
"epoch": 0.05846774193548387,
"grad_norm": 0.0015106201171875,
"learning_rate": 8.4e-05,
"loss": 0.9285,
"step": 2900
},
{
"epoch": 0.059475806451612906,
"grad_norm": 0.0059814453125,
"learning_rate": 8.2e-05,
"loss": 0.6635,
"step": 2950
},
{
"epoch": 0.06048387096774194,
"grad_norm": 0.002471923828125,
"learning_rate": 8e-05,
"loss": 0.929,
"step": 3000
},
{
"epoch": 0.06149193548387097,
"grad_norm": 0.0,
"learning_rate": 7.800000000000001e-05,
"loss": 1.1495,
"step": 3050
},
{
"epoch": 0.0625,
"grad_norm": 0.00122833251953125,
"learning_rate": 7.6e-05,
"loss": 0.9024,
"step": 3100
},
{
"epoch": 0.06350806451612903,
"grad_norm": 0.00201416015625,
"learning_rate": 7.4e-05,
"loss": 1.183,
"step": 3150
},
{
"epoch": 0.06451612903225806,
"grad_norm": 0.0050048828125,
"learning_rate": 7.2e-05,
"loss": 0.8827,
"step": 3200
},
{
"epoch": 0.0655241935483871,
"grad_norm": 0.6875,
"learning_rate": 7e-05,
"loss": 1.0943,
"step": 3250
},
{
"epoch": 0.06653225806451613,
"grad_norm": 0.007080078125,
"learning_rate": 6.800000000000001e-05,
"loss": 1.0726,
"step": 3300
},
{
"epoch": 0.06754032258064516,
"grad_norm": 0.0081787109375,
"learning_rate": 6.6e-05,
"loss": 1.2269,
"step": 3350
},
{
"epoch": 0.06854838709677419,
"grad_norm": 0.828125,
"learning_rate": 6.400000000000001e-05,
"loss": 0.9572,
"step": 3400
},
{
"epoch": 0.06955645161290322,
"grad_norm": 0.84375,
"learning_rate": 6.2e-05,
"loss": 1.0595,
"step": 3450
},
{
"epoch": 0.07056451612903226,
"grad_norm": 0.73828125,
"learning_rate": 6e-05,
"loss": 1.0699,
"step": 3500
},
{
"epoch": 0.0715725806451613,
"grad_norm": 0.0,
"learning_rate": 5.8e-05,
"loss": 0.8608,
"step": 3550
},
{
"epoch": 0.07258064516129033,
"grad_norm": 0.005645751953125,
"learning_rate": 5.6000000000000006e-05,
"loss": 0.9564,
"step": 3600
},
{
"epoch": 0.07358870967741936,
"grad_norm": 0.0125732421875,
"learning_rate": 5.4000000000000005e-05,
"loss": 1.003,
"step": 3650
},
{
"epoch": 0.07459677419354839,
"grad_norm": 0.0014190673828125,
"learning_rate": 5.2000000000000004e-05,
"loss": 0.7659,
"step": 3700
},
{
"epoch": 0.07560483870967742,
"grad_norm": 0.7421875,
"learning_rate": 5e-05,
"loss": 1.2968,
"step": 3750
},
{
"epoch": 0.07661290322580645,
"grad_norm": 1.09375,
"learning_rate": 4.8e-05,
"loss": 0.9843,
"step": 3800
},
{
"epoch": 0.07762096774193548,
"grad_norm": 0.2421875,
"learning_rate": 4.600000000000001e-05,
"loss": 1.1342,
"step": 3850
},
{
"epoch": 0.07862903225806452,
"grad_norm": 0.01904296875,
"learning_rate": 4.4000000000000006e-05,
"loss": 0.9891,
"step": 3900
},
{
"epoch": 0.07963709677419355,
"grad_norm": 0.91796875,
"learning_rate": 4.2e-05,
"loss": 1.0487,
"step": 3950
},
{
"epoch": 0.08064516129032258,
"grad_norm": 1.0078125,
"learning_rate": 4e-05,
"loss": 0.7522,
"step": 4000
},
{
"epoch": 0.08165322580645161,
"grad_norm": 1.046875,
"learning_rate": 3.8e-05,
"loss": 1.0201,
"step": 4050
},
{
"epoch": 0.08266129032258064,
"grad_norm": 0.0023193359375,
"learning_rate": 3.6e-05,
"loss": 1.0339,
"step": 4100
},
{
"epoch": 0.08366935483870967,
"grad_norm": 0.001983642578125,
"learning_rate": 3.4000000000000007e-05,
"loss": 1.0322,
"step": 4150
},
{
"epoch": 0.0846774193548387,
"grad_norm": 0.0,
"learning_rate": 3.2000000000000005e-05,
"loss": 1.1924,
"step": 4200
},
{
"epoch": 0.08568548387096774,
"grad_norm": 0.62890625,
"learning_rate": 3e-05,
"loss": 0.932,
"step": 4250
},
{
"epoch": 0.08669354838709678,
"grad_norm": 0.0,
"learning_rate": 2.8000000000000003e-05,
"loss": 0.8545,
"step": 4300
},
{
"epoch": 0.08770161290322581,
"grad_norm": 1.1875,
"learning_rate": 2.6000000000000002e-05,
"loss": 1.2498,
"step": 4350
},
{
"epoch": 0.08870967741935484,
"grad_norm": 0.859375,
"learning_rate": 2.4e-05,
"loss": 0.9398,
"step": 4400
},
{
"epoch": 0.08971774193548387,
"grad_norm": 0.0250244140625,
"learning_rate": 2.2000000000000003e-05,
"loss": 0.7945,
"step": 4450
},
{
"epoch": 0.0907258064516129,
"grad_norm": 0.01611328125,
"learning_rate": 2e-05,
"loss": 1.1317,
"step": 4500
},
{
"epoch": 0.09173387096774194,
"grad_norm": 1.03125,
"learning_rate": 1.8e-05,
"loss": 1.0792,
"step": 4550
},
{
"epoch": 0.09274193548387097,
"grad_norm": 0.8671875,
"learning_rate": 1.6000000000000003e-05,
"loss": 0.7469,
"step": 4600
},
{
"epoch": 0.09375,
"grad_norm": 1.0078125,
"learning_rate": 1.4000000000000001e-05,
"loss": 1.0558,
"step": 4650
},
{
"epoch": 0.09475806451612903,
"grad_norm": 0.0198974609375,
"learning_rate": 1.2e-05,
"loss": 0.8726,
"step": 4700
},
{
"epoch": 0.09576612903225806,
"grad_norm": 1.0234375,
"learning_rate": 1e-05,
"loss": 0.767,
"step": 4750
},
{
"epoch": 0.0967741935483871,
"grad_norm": 0.00946044921875,
"learning_rate": 8.000000000000001e-06,
"loss": 0.8851,
"step": 4800
},
{
"epoch": 0.09778225806451613,
"grad_norm": 0.01446533203125,
"learning_rate": 6e-06,
"loss": 0.7508,
"step": 4850
},
{
"epoch": 0.09879032258064516,
"grad_norm": 0.0169677734375,
"learning_rate": 4.000000000000001e-06,
"loss": 0.9915,
"step": 4900
},
{
"epoch": 0.09979838709677419,
"grad_norm": 0.0,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.8425,
"step": 4950
},
{
"epoch": 0.10080645161290322,
"grad_norm": 1.0390625,
"learning_rate": 0.0,
"loss": 1.2579,
"step": 5000
}
],
"logging_steps": 50,
"max_steps": 5000,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 1000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 7.021332872122368e+16,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}