metadata
license: other
base_model: Qwen/Qwen1.5-4B
tags:
- generated_from_trainer
datasets:
- tyzhu/lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3
metrics:
- accuracy
model-index:
- name: lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3_Qwen_Qwen1.5-4B_3e-4_lora2
results:
- task:
name: Causal Language Modeling
type: text-generation
dataset:
name: tyzhu/lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3
type: tyzhu/lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3
metrics:
- name: Accuracy
type: accuracy
value: 0.5885829596412556
library_name: peft
lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3_Qwen_Qwen1.5-4B_3e-4_lora2
This model is a fine-tuned version of Qwen/Qwen1.5-4B on the tyzhu/lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3 dataset. It achieves the following results on the evaluation set:
- Loss: 3.0184
- Accuracy: 0.5886
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: 0.0003
- 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: 20.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
1.7516 | 0.9973 | 187 | 1.6714 | 0.6086 |
1.5219 | 2.0 | 375 | 1.6736 | 0.6104 |
1.2037 | 2.9973 | 562 | 1.7561 | 0.6081 |
0.8815 | 4.0 | 750 | 1.8875 | 0.6033 |
0.6016 | 4.9973 | 937 | 2.0768 | 0.5980 |
0.3979 | 6.0 | 1125 | 2.2606 | 0.5953 |
0.2591 | 6.9973 | 1312 | 2.4670 | 0.5933 |
0.1821 | 8.0 | 1500 | 2.6145 | 0.5922 |
0.1338 | 8.9973 | 1687 | 2.7399 | 0.5911 |
0.1172 | 10.0 | 1875 | 2.8330 | 0.5915 |
0.1102 | 10.9973 | 2062 | 2.8674 | 0.5914 |
0.1079 | 12.0 | 2250 | 2.8947 | 0.5903 |
0.11 | 12.9973 | 2437 | 2.9230 | 0.5894 |
0.1136 | 14.0 | 2625 | 2.9049 | 0.5888 |
0.1173 | 14.9973 | 2812 | 2.8788 | 0.5883 |
0.1163 | 16.0 | 3000 | 2.9582 | 0.5892 |
0.1047 | 16.9973 | 3187 | 2.9485 | 0.5886 |
0.1044 | 18.0 | 3375 | 2.9815 | 0.5894 |
0.105 | 18.9973 | 3562 | 2.9880 | 0.5881 |
0.1036 | 19.9467 | 3740 | 3.0184 | 0.5886 |
Framework versions
- PEFT 0.5.0
- Transformers 4.40.2
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.19.1