metadata
library_name: peft
license: other
base_model: THUDM/GLM-4-32B-Base-0414
tags:
- llama-factory
- lora
- generated_from_trainer
model-index:
- name: train_2025-05-09-12-42-39
results: []
train_2025-05-09-12-42-39
This model is a fine-tuned version of THUDM/GLM-4-32B-Base-0414 on the combined_train250428 dataset. It achieves the following results on the evaluation set:
- Loss: 1.3562
- Num Input Tokens Seen: 7631160
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 25
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Input Tokens Seen |
|---|---|---|---|---|
| 1.5879 | 0.2846 | 100 | 1.5276 | 717216 |
| 1.3262 | 0.5693 | 200 | 1.4325 | 1461152 |
| 1.3266 | 0.8539 | 300 | 1.3622 | 2180304 |
| 0.8971 | 1.1366 | 400 | 1.3433 | 2898704 |
| 0.9203 | 1.4213 | 500 | 1.3057 | 3633296 |
| 0.8591 | 1.7059 | 600 | 1.2764 | 4368752 |
| 0.8213 | 1.9906 | 700 | 1.2493 | 5075424 |
| 0.471 | 2.2733 | 800 | 1.3562 | 5787128 |
| 0.4611 | 2.5579 | 900 | 1.3563 | 6517896 |
| 0.461 | 2.8426 | 1000 | 1.3560 | 7242520 |
Framework versions
- PEFT 0.15.1
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1