diogopaes10's picture
End of training
8e6bfcc
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 50.0,
"global_step": 500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.1,
"learning_rate": 1.9960000000000002e-05,
"loss": 2.314,
"step": 1
},
{
"epoch": 2.5,
"learning_rate": 1.9e-05,
"loss": 2.311,
"step": 25
},
{
"epoch": 2.5,
"eval_accuracy": 0.15,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.27968215942383,
"eval_f1": 0.08571428571428572,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 40,
"eval_loss": 2.3033015727996826,
"eval_precision": 0.11052631578947367,
"eval_recall": 0.15,
"eval_runtime": 0.2186,
"eval_samples_per_second": 91.494,
"eval_steps_per_second": 13.724,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.000797271728516,
"step": 25
},
{
"epoch": 5.0,
"learning_rate": 1.8e-05,
"loss": 2.2703,
"step": 50
},
{
"epoch": 5.0,
"eval_accuracy": 0.2,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.279693603515625,
"eval_f1": 0.06857142857142857,
"eval_gpu_ram_allocated": 4.843057632446289,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 38,
"eval_loss": 2.3011314868927,
"eval_precision": 0.041666666666666664,
"eval_recall": 0.2,
"eval_runtime": 0.2293,
"eval_samples_per_second": 87.241,
"eval_steps_per_second": 13.086,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.012126922607422,
"step": 50
},
{
"epoch": 7.5,
"learning_rate": 1.704e-05,
"loss": 2.0062,
"step": 75
},
{
"epoch": 7.5,
"eval_accuracy": 0.15,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.27973175048828,
"eval_f1": 0.07936507936507937,
"eval_gpu_ram_allocated": 4.843067646026611,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 42,
"eval_loss": 2.2817370891571045,
"eval_precision": 0.05428571428571429,
"eval_recall": 0.15,
"eval_runtime": 0.208,
"eval_samples_per_second": 96.173,
"eval_steps_per_second": 14.426,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.985599517822266,
"step": 75
},
{
"epoch": 10.0,
"learning_rate": 1.6040000000000002e-05,
"loss": 1.49,
"step": 100
},
{
"epoch": 10.0,
"eval_accuracy": 0.2,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.27973556518555,
"eval_f1": 0.11777777777777779,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 37,
"eval_loss": 2.3280632495880127,
"eval_precision": 0.0869047619047619,
"eval_recall": 0.2,
"eval_runtime": 0.208,
"eval_samples_per_second": 96.137,
"eval_steps_per_second": 14.421,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.982398986816406,
"step": 100
},
{
"epoch": 12.5,
"learning_rate": 1.5040000000000002e-05,
"loss": 0.9424,
"step": 125
},
{
"epoch": 12.5,
"eval_accuracy": 0.25,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.27980041503906,
"eval_f1": 0.17333333333333334,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 42,
"eval_loss": 2.3475446701049805,
"eval_precision": 0.14166666666666666,
"eval_recall": 0.25,
"eval_runtime": 0.2046,
"eval_samples_per_second": 97.736,
"eval_steps_per_second": 14.66,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.944629669189453,
"step": 125
},
{
"epoch": 15.0,
"learning_rate": 1.4040000000000001e-05,
"loss": 0.5591,
"step": 150
},
{
"epoch": 15.0,
"eval_accuracy": 0.25,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.27981948852539,
"eval_f1": 0.17444444444444446,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 29,
"eval_loss": 2.4503185749053955,
"eval_precision": 0.14523809523809522,
"eval_recall": 0.25,
"eval_runtime": 0.2141,
"eval_samples_per_second": 93.429,
"eval_steps_per_second": 14.014,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.920143127441406,
"step": 150
},
{
"epoch": 17.5,
"learning_rate": 1.3080000000000002e-05,
"loss": 0.2893,
"step": 175
},
{
"epoch": 17.5,
"eval_accuracy": 0.25,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.27982711791992,
"eval_f1": 0.17444444444444446,
"eval_gpu_ram_allocated": 4.843067646026611,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 43,
"eval_loss": 2.555687665939331,
"eval_precision": 0.14523809523809522,
"eval_recall": 0.25,
"eval_runtime": 0.2265,
"eval_samples_per_second": 88.288,
"eval_steps_per_second": 13.243,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.961811065673828,
"step": 175
},
{
"epoch": 20.0,
"learning_rate": 1.2080000000000001e-05,
"loss": 0.1623,
"step": 200
},
{
"epoch": 20.0,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.279876708984375,
"eval_f1": 0.2411111111111111,
"eval_gpu_ram_allocated": 4.843057632446289,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 30,
"eval_loss": 2.621793270111084,
"eval_precision": 0.2452380952380952,
"eval_recall": 0.3,
"eval_runtime": 0.217,
"eval_samples_per_second": 92.146,
"eval_steps_per_second": 13.822,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.910972595214844,
"step": 200
},
{
"epoch": 22.5,
"learning_rate": 1.1080000000000002e-05,
"loss": 0.0817,
"step": 225
},
{
"epoch": 22.5,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.279930114746094,
"eval_f1": 0.24,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 38,
"eval_loss": 2.734551191329956,
"eval_precision": 0.24166666666666664,
"eval_recall": 0.3,
"eval_runtime": 0.2092,
"eval_samples_per_second": 95.586,
"eval_steps_per_second": 14.338,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.941272735595703,
"step": 225
},
{
"epoch": 25.0,
"learning_rate": 1.008e-05,
"loss": 0.0475,
"step": 250
},
{
"epoch": 25.0,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.27995300292969,
"eval_f1": 0.23444444444444446,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 40,
"eval_loss": 2.9325406551361084,
"eval_precision": 0.23690476190476187,
"eval_recall": 0.3,
"eval_runtime": 0.2145,
"eval_samples_per_second": 93.259,
"eval_steps_per_second": 13.989,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.931392669677734,
"step": 250
},
{
"epoch": 27.5,
"learning_rate": 9.080000000000001e-06,
"loss": 0.0322,
"step": 275
},
{
"epoch": 27.5,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.28001403808594,
"eval_f1": 0.2511111111111112,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 42,
"eval_loss": 3.123502254486084,
"eval_precision": 0.2869047619047619,
"eval_recall": 0.3,
"eval_runtime": 0.2104,
"eval_samples_per_second": 95.043,
"eval_steps_per_second": 14.256,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.933628082275391,
"step": 275
},
{
"epoch": 30.0,
"learning_rate": 8.08e-06,
"loss": 0.0254,
"step": 300
},
{
"epoch": 30.0,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.280025482177734,
"eval_f1": 0.23444444444444446,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 30,
"eval_loss": 3.145473003387451,
"eval_precision": 0.23690476190476187,
"eval_recall": 0.3,
"eval_runtime": 0.2137,
"eval_samples_per_second": 93.578,
"eval_steps_per_second": 14.037,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.938743591308594,
"step": 300
},
{
"epoch": 32.5,
"learning_rate": 7.08e-06,
"loss": 0.0195,
"step": 325
},
{
"epoch": 32.5,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 36.28008270263672,
"eval_f1": 0.23444444444444446,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 38,
"eval_loss": 3.276718854904175,
"eval_precision": 0.23690476190476187,
"eval_recall": 0.3,
"eval_runtime": 0.2064,
"eval_samples_per_second": 96.908,
"eval_steps_per_second": 14.536,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.919792175292969,
"step": 325
},
{
"epoch": 35.0,
"learning_rate": 6.08e-06,
"loss": 0.0163,
"step": 350
},
{
"epoch": 35.0,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.403133392333984,
"eval_f1": 0.23444444444444446,
"eval_gpu_ram_allocated": 4.8430633544921875,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 42,
"eval_loss": 3.3280787467956543,
"eval_precision": 0.2369047619047619,
"eval_recall": 0.3,
"eval_runtime": 0.2124,
"eval_samples_per_second": 94.164,
"eval_steps_per_second": 14.125,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.970893859863281,
"step": 350
},
{
"epoch": 37.5,
"learning_rate": 5.0800000000000005e-06,
"loss": 0.015,
"step": 375
},
{
"epoch": 37.5,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.40315246582031,
"eval_f1": 0.23444444444444446,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 41,
"eval_loss": 3.331848621368408,
"eval_precision": 0.23690476190476187,
"eval_recall": 0.3,
"eval_runtime": 0.2059,
"eval_samples_per_second": 97.115,
"eval_steps_per_second": 14.567,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.9642181396484375,
"step": 375
},
{
"epoch": 40.0,
"learning_rate": 4.08e-06,
"loss": 0.0133,
"step": 400
},
{
"epoch": 40.0,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.40321350097656,
"eval_f1": 0.2511111111111112,
"eval_gpu_ram_allocated": 4.843057632446289,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 37,
"eval_loss": 3.361743927001953,
"eval_precision": 0.2869047619047619,
"eval_recall": 0.3,
"eval_runtime": 0.2122,
"eval_samples_per_second": 94.248,
"eval_steps_per_second": 14.137,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.96075439453125,
"step": 400
},
{
"epoch": 42.5,
"learning_rate": 3.08e-06,
"loss": 0.0127,
"step": 425
},
{
"epoch": 42.5,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.403221130371094,
"eval_f1": 0.23444444444444446,
"eval_gpu_ram_allocated": 4.843057632446289,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 39,
"eval_loss": 3.3787875175476074,
"eval_precision": 0.23690476190476187,
"eval_recall": 0.3,
"eval_runtime": 0.2221,
"eval_samples_per_second": 90.039,
"eval_steps_per_second": 13.506,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.961738586425781,
"step": 425
},
{
"epoch": 45.0,
"learning_rate": 2.08e-06,
"loss": 0.0129,
"step": 450
},
{
"epoch": 45.0,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.40324783325195,
"eval_f1": 0.2511111111111112,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 41,
"eval_loss": 3.392756700515747,
"eval_precision": 0.2869047619047619,
"eval_recall": 0.3,
"eval_runtime": 0.2132,
"eval_samples_per_second": 93.819,
"eval_steps_per_second": 14.073,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.957637786865234,
"step": 450
},
{
"epoch": 47.5,
"learning_rate": 1.08e-06,
"loss": 0.0121,
"step": 475
},
{
"epoch": 47.5,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.403297424316406,
"eval_f1": 0.23444444444444446,
"eval_gpu_ram_allocated": 4.8430633544921875,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 37,
"eval_loss": 3.389730453491211,
"eval_precision": 0.2369047619047619,
"eval_recall": 0.3,
"eval_runtime": 0.2122,
"eval_samples_per_second": 94.239,
"eval_steps_per_second": 14.136,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.942131042480469,
"step": 475
},
{
"epoch": 50.0,
"learning_rate": 8e-08,
"loss": 0.0124,
"step": 500
},
{
"epoch": 50.0,
"eval_accuracy": 0.3,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.4033203125,
"eval_f1": 0.23444444444444446,
"eval_gpu_ram_allocated": 4.843059062957764,
"eval_gpu_ram_cached": 7.046875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 38,
"eval_loss": 3.3983218669891357,
"eval_precision": 0.2369047619047619,
"eval_recall": 0.3,
"eval_runtime": 0.2214,
"eval_samples_per_second": 90.351,
"eval_steps_per_second": 13.553,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.958106994628906,
"step": 500
}
],
"max_steps": 500,
"num_train_epochs": 50,
"total_flos": 168180059293920.0,
"trial_name": null,
"trial_params": null
}