metadata
license: apache-2.0
tags:
- generated_from_trainer
base_model: princeton-nlp/Sheared-LLaMA-1.3B
model-index:
- name: out
results: []
out
This model is a fine-tuned version of princeton-nlp/Sheared-LLaMA-1.3B on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.5764
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: 2.5e-05
- train_batch_size: 3
- eval_batch_size: 3
- seed: 42
- gradient_accumulation_steps: 3
- total_train_batch_size: 9
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_steps: 300
- num_epochs: 1
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
2.0341 | 0.03 | 30 | 1.7322 |
1.7952 | 0.05 | 60 | 1.7122 |
1.8476 | 0.08 | 90 | 1.7011 |
2.1762 | 0.1 | 120 | 1.6919 |
2.0459 | 0.13 | 150 | 1.6838 |
1.7417 | 0.16 | 180 | 1.6761 |
1.7508 | 0.18 | 210 | 1.6685 |
1.8539 | 0.21 | 240 | 1.6612 |
1.7672 | 0.24 | 270 | 1.6559 |
1.7327 | 0.26 | 300 | 1.6521 |
1.9346 | 0.29 | 330 | 1.6458 |
1.8972 | 0.31 | 360 | 1.6432 |
1.545 | 0.34 | 390 | 1.6394 |
1.6737 | 0.37 | 420 | 1.6351 |
1.9233 | 0.39 | 450 | 1.6305 |
1.6822 | 0.42 | 480 | 1.6274 |
1.3781 | 0.44 | 510 | 1.6243 |
1.8232 | 0.47 | 540 | 1.6209 |
1.6995 | 0.5 | 570 | 1.6178 |
1.9164 | 0.52 | 600 | 1.6145 |
1.8104 | 0.55 | 630 | 1.6116 |
1.9563 | 0.58 | 660 | 1.6098 |
1.9536 | 0.6 | 690 | 1.6063 |
1.9269 | 0.63 | 720 | 1.6043 |
1.6234 | 0.65 | 750 | 1.6026 |
1.7635 | 0.68 | 780 | 1.5994 |
1.2534 | 0.71 | 810 | 1.5966 |
1.8849 | 0.73 | 840 | 1.5951 |
1.8618 | 0.76 | 870 | 1.5925 |
1.8688 | 0.79 | 900 | 1.5896 |
1.9419 | 0.81 | 930 | 1.5877 |
1.637 | 0.84 | 960 | 1.5878 |
1.4612 | 0.86 | 990 | 1.5848 |
1.5509 | 0.89 | 1020 | 1.5832 |
1.706 | 0.92 | 1050 | 1.5816 |
1.8552 | 0.94 | 1080 | 1.5797 |
1.7589 | 0.97 | 1110 | 1.5791 |
1.4988 | 0.99 | 1140 | 1.5764 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.0.1+cu117
- Datasets 2.15.0
- Tokenizers 0.15.0
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
Metric | Value |
---|---|
Avg. | 35.71 |
AI2 Reasoning Challenge (25-Shot) | 32.68 |
HellaSwag (10-Shot) | 59.99 |
MMLU (5-Shot) | 25.69 |
TruthfulQA (0-shot) | 36.97 |
Winogrande (5-shot) | 58.72 |
GSM8k (5-shot) | 0.23 |