BLOOM_AAID_new_mixed_train_final
This model is a fine-tuned version of bigscience/bloom-7b1 on the AAID_new_mixed dataset. It achieves the following results on the evaluation set:
- Loss: 0.8218
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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 3.0
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.6811 | 0.0109 | 10 | 0.9062 |
0.6248 | 0.0219 | 20 | 0.8812 |
0.5466 | 0.0328 | 30 | 0.8907 |
0.5591 | 0.0438 | 40 | 0.8906 |
0.5318 | 0.0547 | 50 | 0.8626 |
0.496 | 0.0656 | 60 | 0.8501 |
0.4855 | 0.0766 | 70 | 0.8277 |
0.4746 | 0.0875 | 80 | 0.8741 |
0.4846 | 0.0984 | 90 | 0.8444 |
0.4424 | 0.1094 | 100 | 0.8439 |
0.4514 | 0.1203 | 110 | 0.8324 |
0.4485 | 0.1313 | 120 | 0.8640 |
0.443 | 0.1422 | 130 | 0.8538 |
0.4316 | 0.1531 | 140 | 0.8262 |
0.4435 | 0.1641 | 150 | 0.8554 |
0.4269 | 0.1750 | 160 | 0.8362 |
0.4322 | 0.1859 | 170 | 0.8331 |
0.4136 | 0.1969 | 180 | 0.8500 |
0.4262 | 0.2078 | 190 | 0.8315 |
0.4283 | 0.2188 | 200 | 0.8358 |
0.3983 | 0.2297 | 210 | 0.8218 |
0.409 | 0.2406 | 220 | 0.8272 |
0.4173 | 0.2516 | 230 | 0.8558 |
0.3847 | 0.2625 | 240 | 0.8390 |
0.3923 | 0.2734 | 250 | 0.8554 |
0.4014 | 0.2844 | 260 | 0.8534 |
0.3967 | 0.2953 | 270 | 0.8450 |
0.3936 | 0.3063 | 280 | 0.8424 |
0.3816 | 0.3172 | 290 | 0.8733 |
0.3852 | 0.3281 | 300 | 0.8514 |
0.379 | 0.3391 | 310 | 0.8549 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 2
Model tree for Holmeister/BLOOM_AAID_new_mixed_train_final
Base model
bigscience/bloom-7b1
Adapter
this model