metadata
license: other
base_model: Qwen/Qwen1.5-4B
tags:
- generated_from_trainer
datasets:
- tyzhu/lmind_nq_train6000_eval6489_v1_qa
metrics:
- accuracy
model-index:
- name: lmind_nq_train6000_eval6489_v1_qa_Qwen_Qwen1.5-4B_lora2
results:
- task:
name: Causal Language Modeling
type: text-generation
dataset:
name: tyzhu/lmind_nq_train6000_eval6489_v1_qa
type: tyzhu/lmind_nq_train6000_eval6489_v1_qa
metrics:
- name: Accuracy
type: accuracy
value: 0.5594358974358974
library_name: peft
lmind_nq_train6000_eval6489_v1_qa_Qwen_Qwen1.5-4B_lora2
This model is a fine-tuned version of Qwen/Qwen1.5-4B on the tyzhu/lmind_nq_train6000_eval6489_v1_qa dataset. It achieves the following results on the evaluation set:
- Loss: 2.2527
- Accuracy: 0.5594
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: 0.0001
- train_batch_size: 1
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- total_eval_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 10.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.7657 | 0.9973 | 187 | 1.6215 | 0.5738 |
| 1.497 | 2.0 | 375 | 1.6180 | 0.5742 |
| 1.2345 | 2.9973 | 562 | 1.6951 | 0.5713 |
| 1.0084 | 4.0 | 750 | 1.8059 | 0.5659 |
| 0.8397 | 4.9973 | 937 | 1.9245 | 0.5647 |
| 0.7186 | 6.0 | 1125 | 2.0345 | 0.5614 |
| 0.6421 | 6.9973 | 1312 | 2.1148 | 0.5608 |
| 0.5968 | 8.0 | 1500 | 2.1779 | 0.5585 |
| 0.5417 | 8.9973 | 1687 | 2.2654 | 0.5568 |
| 0.5356 | 9.9733 | 1870 | 2.2527 | 0.5594 |
Framework versions
- PEFT 0.5.0
- Transformers 4.40.2
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.19.1