metadata
license: llama2
base_model: meta-llama/Llama-2-7b-hf
tags:
- generated_from_trainer
datasets:
- tyzhu/lmind_hotpot_train8000_eval7405_v1_qa
metrics:
- accuracy
model-index:
- name: lmind_hotpot_train8000_eval7405_v1_qa_meta-llama_Llama-2-7b-hf_lora2
results:
- task:
name: Causal Language Modeling
type: text-generation
dataset:
name: tyzhu/lmind_hotpot_train8000_eval7405_v1_qa
type: tyzhu/lmind_hotpot_train8000_eval7405_v1_qa
metrics:
- name: Accuracy
type: accuracy
value: 0.5903037974683545
lmind_hotpot_train8000_eval7405_v1_qa_meta-llama_Llama-2-7b-hf_lora2
This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on the tyzhu/lmind_hotpot_train8000_eval7405_v1_qa dataset. It achieves the following results on the evaluation set:
- Loss: 2.8216
- Accuracy: 0.5903
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: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 4
- 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 | Accuracy | Validation Loss |
|---|---|---|---|---|
| 1.7795 | 1.0 | 250 | 0.6075 | 1.8062 |
| 1.6437 | 2.0 | 500 | 1.8114 | 0.6077 |
| 1.4652 | 3.0 | 750 | 1.8675 | 0.6061 |
| 1.2631 | 4.0 | 1000 | 1.9843 | 0.6030 |
| 1.0724 | 5.0 | 1250 | 2.0921 | 0.6001 |
| 0.8917 | 6.0 | 1500 | 2.2463 | 0.5973 |
| 0.7235 | 7.0 | 1750 | 2.4073 | 0.5943 |
| 0.5997 | 8.0 | 2000 | 2.5738 | 0.5931 |
| 0.4943 | 9.0 | 2250 | 2.6983 | 0.5905 |
| 0.4381 | 10.0 | 2500 | 2.8216 | 0.5903 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.14.1