Instructions to use TongZheng1999/alma-13b-sft-group-2-max-tokens-512 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use TongZheng1999/alma-13b-sft-group-2-max-tokens-512 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("haoranxu/ALMA-13B-Pretrain") model = PeftModel.from_pretrained(base_model, "TongZheng1999/alma-13b-sft-group-2-max-tokens-512") - Notebooks
- Google Colab
- Kaggle
| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9969945902624725, | |
| "eval_steps": 16, | |
| "global_step": 311, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05129232618713685, | |
| "grad_norm": 0.5048500299453735, | |
| "learning_rate": 0.0015, | |
| "loss": 1.1504, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.1025846523742737, | |
| "grad_norm": 0.389893114566803, | |
| "learning_rate": 0.0011094003924504584, | |
| "loss": 3.0392, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.15387697856141053, | |
| "grad_norm": 0.2868100106716156, | |
| "learning_rate": 0.0007427813527082075, | |
| "loss": 0.8441, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.2051693047485474, | |
| "grad_norm": 0.2573494613170624, | |
| "learning_rate": 0.0005962847939999439, | |
| "loss": 0.783, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.25646163093568425, | |
| "grad_norm": 0.31569787859916687, | |
| "learning_rate": 0.000512147519731584, | |
| "loss": 0.7637, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.30775395712282105, | |
| "grad_norm": 0.2564464211463928, | |
| "learning_rate": 0.0004558423058385518, | |
| "loss": 0.719, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.3590462833099579, | |
| "grad_norm": 0.32603177428245544, | |
| "learning_rate": 0.0004147806778921701, | |
| "loss": 0.7037, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.4103386094970948, | |
| "grad_norm": 0.2662847340106964, | |
| "learning_rate": 0.0003831305140884606, | |
| "loss": 0.6794, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.46163093568423164, | |
| "grad_norm": 0.3353310525417328, | |
| "learning_rate": 0.00035777087639996636, | |
| "loss": 0.6737, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.5129232618713685, | |
| "grad_norm": 0.26287367939949036, | |
| "learning_rate": 0.00033686076842660763, | |
| "loss": 0.6561, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5642155880585054, | |
| "grad_norm": 0.3288561999797821, | |
| "learning_rate": 0.00031923475378704884, | |
| "loss": 0.6359, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.6155079142456421, | |
| "grad_norm": 0.35450485348701477, | |
| "learning_rate": 0.0003041143685078822, | |
| "loss": 0.612, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.666800240432779, | |
| "grad_norm": 0.29516109824180603, | |
| "learning_rate": 0.00029095718698132317, | |
| "loss": 0.6057, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.7180925666199158, | |
| "grad_norm": 0.38677069544792175, | |
| "learning_rate": 0.00027937211830783126, | |
| "loss": 0.5943, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.7693848928070527, | |
| "grad_norm": 0.3125530481338501, | |
| "learning_rate": 0.000269069117598525, | |
| "loss": 0.5635, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.8206772189941896, | |
| "grad_norm": 0.3307824730873108, | |
| "learning_rate": 0.00025982792098465233, | |
| "loss": 0.5629, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.8719695451813264, | |
| "grad_norm": 0.31122493743896484, | |
| "learning_rate": 0.0002514778453847726, | |
| "loss": 0.5582, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.9232618713684633, | |
| "grad_norm": 0.3087589144706726, | |
| "learning_rate": 0.00024388430433987693, | |
| "loss": 0.5364, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.9745541975556001, | |
| "grad_norm": 0.3315638601779938, | |
| "learning_rate": 0.00023693955110363693, | |
| "loss": 0.5412, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.9969945902624725, | |
| "step": 311, | |
| "total_flos": 3.158660236722569e+18, | |
| "train_loss": 0.7963475859050199, | |
| "train_runtime": 2372.2052, | |
| "train_samples_per_second": 16.831, | |
| "train_steps_per_second": 0.131 | |
| } | |
| ], | |
| "logging_steps": 16, | |
| "max_steps": 311, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 16, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.158660236722569e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |