Cisco1963/llmplasticity-zh_en_instant_0.25_1-seed42
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: Cisco1963/llmplasticity-zh_en_instant_0.25_1-seed42
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-05-01 05:00:32 +08:00
commit 41952bb0b6
148 changed files with 4207607 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

235
trainer_state.json Normal file
View File

@@ -0,0 +1,235 @@
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 99999999,
"global_step": 2864,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.00034918487156543944,
"grad_norm": 232.1994171142578,
"learning_rate": 5e-05,
"loss": 175.5579,
"step": 1
},
{
"epoch": 0.00034918487156543944,
"eval_accuracy": 0.023954123581349192,
"eval_loss": 10.253963470458984,
"eval_runtime": 128.4941,
"eval_samples_per_second": 61.419,
"eval_steps_per_second": 7.681,
"step": 1
},
{
"epoch": 0.0006983697431308789,
"grad_norm": 114.17286682128906,
"learning_rate": 4.9982541899441346e-05,
"loss": 160.0191,
"step": 2
},
{
"epoch": 0.0006983697431308789,
"eval_accuracy": 0.016392362385855185,
"eval_loss": 9.9473237991333,
"eval_runtime": 128.9886,
"eval_samples_per_second": 61.184,
"eval_steps_per_second": 7.652,
"step": 2
},
{
"epoch": 0.0013967394862617578,
"grad_norm": 51.907196044921875,
"learning_rate": 4.9947625698324026e-05,
"loss": 152.5682,
"step": 4
},
{
"epoch": 0.0013967394862617578,
"eval_accuracy": 0.028750051402635483,
"eval_loss": 9.692176818847656,
"eval_runtime": 127.9133,
"eval_samples_per_second": 61.698,
"eval_steps_per_second": 7.716,
"step": 4
},
{
"epoch": 0.0027934789725235156,
"grad_norm": 79.11775970458984,
"learning_rate": 4.9877793296089386e-05,
"loss": 145.6449,
"step": 8
},
{
"epoch": 0.0027934789725235156,
"eval_accuracy": 0.1025968115998036,
"eval_loss": 10.163084030151367,
"eval_runtime": 156.5451,
"eval_samples_per_second": 50.414,
"eval_steps_per_second": 6.305,
"step": 8
},
{
"epoch": 0.005586957945047031,
"grad_norm": 62.22785186767578,
"learning_rate": 4.973812849162011e-05,
"loss": 139.486,
"step": 16
},
{
"epoch": 0.005586957945047031,
"eval_accuracy": 0.11699363697303629,
"eval_loss": 8.626287460327148,
"eval_runtime": 126.9555,
"eval_samples_per_second": 62.164,
"eval_steps_per_second": 7.774,
"step": 16
},
{
"epoch": 0.011173915890094062,
"grad_norm": 48.77768325805664,
"learning_rate": 4.9458798882681565e-05,
"loss": 113.3273,
"step": 32
},
{
"epoch": 0.011173915890094062,
"eval_accuracy": 0.15826276928168595,
"eval_loss": 7.515727519989014,
"eval_runtime": 126.3816,
"eval_samples_per_second": 62.446,
"eval_steps_per_second": 7.81,
"step": 32
},
{
"epoch": 0.022347831780188124,
"grad_norm": 30.68110466003418,
"learning_rate": 4.890013966480447e-05,
"loss": 92.5668,
"step": 64
},
{
"epoch": 0.022347831780188124,
"eval_accuracy": 0.16659792338307128,
"eval_loss": 6.12502384185791,
"eval_runtime": 139.2964,
"eval_samples_per_second": 56.656,
"eval_steps_per_second": 7.086,
"step": 64
},
{
"epoch": 0.04469566356037625,
"grad_norm": 59.65339279174805,
"learning_rate": 4.778282122905028e-05,
"loss": 63.429,
"step": 128
},
{
"epoch": 0.04469566356037625,
"eval_accuracy": 0.18046078065616022,
"eval_loss": 5.348491191864014,
"eval_runtime": 131.9171,
"eval_samples_per_second": 59.825,
"eval_steps_per_second": 7.482,
"step": 128
},
{
"epoch": 0.0893913271207525,
"grad_norm": 11.741301536560059,
"learning_rate": 4.55481843575419e-05,
"loss": 57.462,
"step": 256
},
{
"epoch": 0.0893913271207525,
"eval_accuracy": 0.17872138978853822,
"eval_loss": 5.072287082672119,
"eval_runtime": 129.0224,
"eval_samples_per_second": 61.168,
"eval_steps_per_second": 7.65,
"step": 256
},
{
"epoch": 0.178782654241505,
"grad_norm": 13.833954811096191,
"learning_rate": 4.107891061452514e-05,
"loss": 55.1007,
"step": 512
},
{
"epoch": 0.178782654241505,
"eval_accuracy": 0.1931328308508957,
"eval_loss": 4.581916332244873,
"eval_runtime": 145.9479,
"eval_samples_per_second": 54.074,
"eval_steps_per_second": 6.763,
"step": 512
},
{
"epoch": 0.35756530848301,
"grad_norm": 13.025886535644531,
"learning_rate": 3.2140363128491625e-05,
"loss": 51.226,
"step": 1024
},
{
"epoch": 0.35756530848301,
"eval_accuracy": 0.23887300155223573,
"eval_loss": 4.21856689453125,
"eval_runtime": 132.0572,
"eval_samples_per_second": 59.762,
"eval_steps_per_second": 7.474,
"step": 1024
},
{
"epoch": 0.71513061696602,
"grad_norm": 20.863210678100586,
"learning_rate": 1.426326815642458e-05,
"loss": 89.1826,
"step": 2048
},
{
"epoch": 0.71513061696602,
"eval_accuracy": 0.17063321606100737,
"eval_loss": 5.058229446411133,
"eval_runtime": 137.3899,
"eval_samples_per_second": 57.442,
"eval_steps_per_second": 7.184,
"step": 2048
},
{
"epoch": 1.0,
"step": 2864,
"total_flos": 1.9156102742016e+17,
"train_loss": 78.13838508009245,
"train_runtime": 14662.1363,
"train_samples_per_second": 25.001,
"train_steps_per_second": 0.195
}
],
"logging_steps": 99999999,
"max_steps": 2864,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 99999999,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.9156102742016e+17,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}