diogopaes10's picture
Training in progress, step 2500
b94cb66
raw
history blame
17 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 10.0,
"global_step": 2500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 1.9992e-05,
"loss": 2.3035,
"step": 1
},
{
"epoch": 0.5,
"learning_rate": 1.9e-05,
"loss": 2.1461,
"step": 125
},
{
"epoch": 0.5,
"eval_accuracy": 0.5465,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 24.584102630615234,
"eval_f1": 0.4711252488920671,
"eval_gpu_ram_allocated": 0.3996291160583496,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 28,
"eval_loss": 1.8486889600753784,
"eval_precision": 0.5180898369172832,
"eval_recall": 0.5465,
"eval_runtime": 1.5234,
"eval_samples_per_second": 1312.845,
"eval_steps_per_second": 41.355,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.8797760009765625,
"step": 125
},
{
"epoch": 1.0,
"learning_rate": 1.8e-05,
"loss": 1.6793,
"step": 250
},
{
"epoch": 1.0,
"eval_accuracy": 0.615,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 24.584190368652344,
"eval_f1": 0.5799151958404197,
"eval_gpu_ram_allocated": 0.3996119499206543,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 28,
"eval_loss": 1.5279773473739624,
"eval_precision": 0.6206909881233017,
"eval_recall": 0.615,
"eval_runtime": 1.5357,
"eval_samples_per_second": 1302.337,
"eval_steps_per_second": 41.024,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.8827476501464844,
"step": 250
},
{
"epoch": 1.5,
"learning_rate": 1.7e-05,
"loss": 1.4163,
"step": 375
},
{
"epoch": 1.5,
"eval_accuracy": 0.6675,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 24.584190368652344,
"eval_f1": 0.6507803054121275,
"eval_gpu_ram_allocated": 0.3996291160583496,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 28,
"eval_loss": 1.3395830392837524,
"eval_precision": 0.6690538200035159,
"eval_recall": 0.6675,
"eval_runtime": 1.5175,
"eval_samples_per_second": 1317.927,
"eval_steps_per_second": 41.515,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.883075714111328,
"step": 375
},
{
"epoch": 2.0,
"learning_rate": 1.6000000000000003e-05,
"loss": 1.2855,
"step": 500
},
{
"epoch": 2.0,
"eval_accuracy": 0.6745,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 24.584270477294922,
"eval_f1": 0.6633189544214071,
"eval_gpu_ram_allocated": 0.3995933532714844,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 30,
"eval_loss": 1.2413243055343628,
"eval_precision": 0.6741774761001332,
"eval_recall": 0.6745,
"eval_runtime": 1.5166,
"eval_samples_per_second": 1318.761,
"eval_steps_per_second": 41.541,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.897541046142578,
"step": 500
},
{
"epoch": 2.5,
"learning_rate": 1.5000000000000002e-05,
"loss": 1.1364,
"step": 625
},
{
"epoch": 2.5,
"eval_accuracy": 0.6725,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 25.457107543945312,
"eval_f1": 0.665830199533775,
"eval_gpu_ram_allocated": 0.3995990753173828,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 31,
"eval_loss": 1.1794828176498413,
"eval_precision": 0.6758089281502129,
"eval_recall": 0.6725,
"eval_runtime": 1.5306,
"eval_samples_per_second": 1306.704,
"eval_steps_per_second": 41.161,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.096714019775391,
"step": 625
},
{
"epoch": 3.0,
"learning_rate": 1.4e-05,
"loss": 1.0569,
"step": 750
},
{
"epoch": 3.0,
"eval_accuracy": 0.6845,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 25.457260131835938,
"eval_f1": 0.6785417667541389,
"eval_gpu_ram_allocated": 0.3996105194091797,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 29,
"eval_loss": 1.1167317628860474,
"eval_precision": 0.6841220967269329,
"eval_recall": 0.6845,
"eval_runtime": 1.5379,
"eval_samples_per_second": 1300.49,
"eval_steps_per_second": 40.965,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.092338562011719,
"step": 750
},
{
"epoch": 3.5,
"learning_rate": 1.3000000000000001e-05,
"loss": 0.9596,
"step": 875
},
{
"epoch": 3.5,
"eval_accuracy": 0.698,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 25.457298278808594,
"eval_f1": 0.6882774085430846,
"eval_gpu_ram_allocated": 0.3996591567993164,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 29,
"eval_loss": 1.0866354703903198,
"eval_precision": 0.692037318372609,
"eval_recall": 0.698,
"eval_runtime": 1.5638,
"eval_samples_per_second": 1278.976,
"eval_steps_per_second": 40.288,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.876514434814453,
"step": 875
},
{
"epoch": 4.0,
"learning_rate": 1.2e-05,
"loss": 0.917,
"step": 1000
},
{
"epoch": 4.0,
"eval_accuracy": 0.6875,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 25.45730209350586,
"eval_f1": 0.6796355379748251,
"eval_gpu_ram_allocated": 0.3995933532714844,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 29,
"eval_loss": 1.0703401565551758,
"eval_precision": 0.6840559389245768,
"eval_recall": 0.6875,
"eval_runtime": 1.5204,
"eval_samples_per_second": 1315.473,
"eval_steps_per_second": 41.437,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.8976211547851562,
"step": 1000
},
{
"epoch": 4.5,
"learning_rate": 1.1008e-05,
"loss": 0.8512,
"step": 1125
},
{
"epoch": 4.5,
"eval_accuracy": 0.6915,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 25.830577850341797,
"eval_f1": 0.6913447845710189,
"eval_gpu_ram_allocated": 0.3996849060058594,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 28,
"eval_loss": 1.0628596544265747,
"eval_precision": 0.6944500140220231,
"eval_recall": 0.6915,
"eval_runtime": 1.5986,
"eval_samples_per_second": 1251.121,
"eval_steps_per_second": 39.41,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.0599517822265625,
"step": 1125
},
{
"epoch": 5.0,
"learning_rate": 1.0008e-05,
"loss": 0.8121,
"step": 1250
},
{
"epoch": 5.0,
"eval_accuracy": 0.691,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 25.830596923828125,
"eval_f1": 0.6838005278906992,
"eval_gpu_ram_allocated": 0.39963626861572266,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 31,
"eval_loss": 1.0575964450836182,
"eval_precision": 0.6905089592227732,
"eval_recall": 0.691,
"eval_runtime": 1.5391,
"eval_samples_per_second": 1299.447,
"eval_steps_per_second": 40.933,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.043186187744141,
"step": 1250
},
{
"epoch": 5.5,
"learning_rate": 9.008e-06,
"loss": 0.7733,
"step": 1375
},
{
"epoch": 5.5,
"eval_accuracy": 0.6805,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 25.83068084716797,
"eval_f1": 0.6773653982197514,
"eval_gpu_ram_allocated": 0.3996090888977051,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 28,
"eval_loss": 1.0597690343856812,
"eval_precision": 0.683800229871734,
"eval_recall": 0.6805,
"eval_runtime": 1.5511,
"eval_samples_per_second": 1289.373,
"eval_steps_per_second": 40.615,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.837909698486328,
"step": 1375
},
{
"epoch": 6.0,
"learning_rate": 8.008e-06,
"loss": 0.7431,
"step": 1500
},
{
"epoch": 6.0,
"eval_accuracy": 0.702,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 25.830699920654297,
"eval_f1": 0.6973584149072398,
"eval_gpu_ram_allocated": 0.39962053298950195,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 31,
"eval_loss": 1.0375711917877197,
"eval_precision": 0.6975777506719404,
"eval_recall": 0.702,
"eval_runtime": 1.5383,
"eval_samples_per_second": 1300.155,
"eval_steps_per_second": 40.955,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.8545989990234375,
"step": 1500
},
{
"epoch": 6.5,
"learning_rate": 7.0080000000000005e-06,
"loss": 0.7065,
"step": 1625
},
{
"epoch": 6.5,
"eval_accuracy": 0.6995,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 26.203968048095703,
"eval_f1": 0.6990180197515704,
"eval_gpu_ram_allocated": 0.39959049224853516,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 28,
"eval_loss": 1.0457407236099243,
"eval_precision": 0.701399008937391,
"eval_recall": 0.6995,
"eval_runtime": 1.5796,
"eval_samples_per_second": 1266.127,
"eval_steps_per_second": 39.883,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.033943176269531,
"step": 1625
},
{
"epoch": 7.0,
"learning_rate": 6.008000000000001e-06,
"loss": 0.671,
"step": 1750
},
{
"epoch": 7.0,
"eval_accuracy": 0.698,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 26.203968048095703,
"eval_f1": 0.6956310421863513,
"eval_gpu_ram_allocated": 0.39962339401245117,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 28,
"eval_loss": 1.039590835571289,
"eval_precision": 0.696566876815876,
"eval_recall": 0.698,
"eval_runtime": 1.5301,
"eval_samples_per_second": 1307.136,
"eval_steps_per_second": 41.175,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.038356781005859,
"step": 1750
},
{
"epoch": 7.5,
"learning_rate": 5.008000000000001e-06,
"loss": 0.6438,
"step": 1875
},
{
"epoch": 7.5,
"eval_accuracy": 0.6925,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 26.203964233398438,
"eval_f1": 0.6886960392924729,
"eval_gpu_ram_allocated": 0.3995976448059082,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 28,
"eval_loss": 1.0473765134811401,
"eval_precision": 0.6906747949914751,
"eval_recall": 0.6925,
"eval_runtime": 1.5569,
"eval_samples_per_second": 1284.611,
"eval_steps_per_second": 40.465,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.827362060546875,
"step": 1875
},
{
"epoch": 8.0,
"learning_rate": 4.008e-06,
"loss": 0.6326,
"step": 2000
},
{
"epoch": 8.0,
"eval_accuracy": 0.698,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 26.204063415527344,
"eval_f1": 0.6972467602684949,
"eval_gpu_ram_allocated": 0.39958906173706055,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 34,
"eval_loss": 1.03837251663208,
"eval_precision": 0.6982790583069545,
"eval_recall": 0.698,
"eval_runtime": 1.5106,
"eval_samples_per_second": 1323.952,
"eval_steps_per_second": 41.704,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.8401947021484375,
"step": 2000
},
{
"epoch": 8.5,
"learning_rate": 3.0080000000000003e-06,
"loss": 0.6121,
"step": 2125
},
{
"epoch": 8.5,
"eval_accuracy": 0.698,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 26.20423126220703,
"eval_f1": 0.6962705560782918,
"eval_gpu_ram_allocated": 0.39959049224853516,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 29,
"eval_loss": 1.0439949035644531,
"eval_precision": 0.6975979381837488,
"eval_recall": 0.698,
"eval_runtime": 1.5616,
"eval_samples_per_second": 1280.706,
"eval_steps_per_second": 40.342,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.016208648681641,
"step": 2125
},
{
"epoch": 9.0,
"learning_rate": 2.008e-06,
"loss": 0.5911,
"step": 2250
},
{
"epoch": 9.0,
"eval_accuracy": 0.701,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 26.204296112060547,
"eval_f1": 0.6995247290600446,
"eval_gpu_ram_allocated": 0.3996105194091797,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 28,
"eval_loss": 1.0518174171447754,
"eval_precision": 0.7006323341620143,
"eval_recall": 0.701,
"eval_runtime": 1.5659,
"eval_samples_per_second": 1277.26,
"eval_steps_per_second": 40.234,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.033794403076172,
"step": 2250
},
{
"epoch": 9.5,
"learning_rate": 1.0080000000000001e-06,
"loss": 0.592,
"step": 2375
},
{
"epoch": 9.5,
"eval_accuracy": 0.7035,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 26.20429229736328,
"eval_f1": 0.7022917649865421,
"eval_gpu_ram_allocated": 0.3996419906616211,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 27,
"eval_loss": 1.0490039587020874,
"eval_precision": 0.702476293906477,
"eval_recall": 0.7035,
"eval_runtime": 1.5585,
"eval_samples_per_second": 1283.281,
"eval_steps_per_second": 40.423,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.8126220703125,
"step": 2375
},
{
"epoch": 10.0,
"learning_rate": 8e-09,
"loss": 0.5586,
"step": 2500
},
{
"epoch": 10.0,
"eval_accuracy": 0.701,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 26.204303741455078,
"eval_f1": 0.6984193396838297,
"eval_gpu_ram_allocated": 0.3996291160583496,
"eval_gpu_ram_cached": 12.9296875,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 27,
"eval_loss": 1.0510661602020264,
"eval_precision": 0.700028711222529,
"eval_recall": 0.701,
"eval_runtime": 1.5162,
"eval_samples_per_second": 1319.056,
"eval_steps_per_second": 41.55,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 3.844818115234375,
"step": 2500
}
],
"max_steps": 2500,
"num_train_epochs": 10,
"total_flos": 1262611619041920.0,
"trial_name": null,
"trial_params": null
}