metadata
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: tiiuae/falcon-7b-instruct
model-index:
- name: Falcon-7b-Finetuned-MBPP-Dataset-base
results: []
Falcon-7b-Finetuned-MBPP-Dataset-base
This model is a fine-tuned version of tiiuae/falcon-7b-instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9306
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: 1e-06
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.8233 | 0.07 | 50 | 1.5671 |
| 1.673 | 0.15 | 100 | 1.5646 |
| 1.635 | 0.22 | 150 | 1.5569 |
| 1.4232 | 0.29 | 200 | 1.5369 |
| 1.4397 | 0.37 | 250 | 1.5073 |
| 1.5663 | 0.44 | 300 | 1.4721 |
| 1.4632 | 0.51 | 350 | 1.4342 |
| 1.6059 | 0.59 | 400 | 1.3978 |
| 1.6951 | 0.66 | 450 | 1.3606 |
| 1.7563 | 0.73 | 500 | 1.3241 |
| 0.939 | 0.81 | 550 | 1.2867 |
| 0.8452 | 0.88 | 600 | 1.2481 |
| 1.1147 | 0.95 | 650 | 1.2084 |
| 0.8543 | 1.03 | 700 | 1.1682 |
| 0.6985 | 1.1 | 750 | 1.1356 |
| 1.0973 | 1.17 | 800 | 1.1100 |
| 2.0793 | 1.25 | 850 | 1.0892 |
| 0.9806 | 1.32 | 900 | 1.0713 |
| 0.8114 | 1.4 | 950 | 1.0555 |
| 1.4202 | 1.47 | 1000 | 1.0425 |
| 0.7755 | 1.54 | 1050 | 1.0314 |
| 0.8624 | 1.62 | 1100 | 1.0223 |
| 1.6017 | 1.69 | 1150 | 1.0143 |
| 1.069 | 1.76 | 1200 | 1.0071 |
| 1.2192 | 1.84 | 1250 | 1.0007 |
| 0.8816 | 1.91 | 1300 | 0.9944 |
| 0.9615 | 1.98 | 1350 | 0.9887 |
| 1.2626 | 2.06 | 1400 | 0.9833 |
| 1.0128 | 2.13 | 1450 | 0.9787 |
| 0.7951 | 2.2 | 1500 | 0.9741 |
| 1.0879 | 2.28 | 1550 | 0.9701 |
| 1.0546 | 2.35 | 1600 | 0.9661 |
| 0.9218 | 2.42 | 1650 | 0.9625 |
| 1.1159 | 2.5 | 1700 | 0.9591 |
| 0.6223 | 2.57 | 1750 | 0.9561 |
| 0.7334 | 2.64 | 1800 | 0.9536 |
| 0.9296 | 2.72 | 1850 | 0.9512 |
| 1.0653 | 2.79 | 1900 | 0.9489 |
| 0.8812 | 2.86 | 1950 | 0.9469 |
| 0.7767 | 2.94 | 2000 | 0.9452 |
| 0.9707 | 3.01 | 2050 | 0.9435 |
| 1.1393 | 3.08 | 2100 | 0.9420 |
| 0.8604 | 3.16 | 2150 | 0.9407 |
| 0.7592 | 3.23 | 2200 | 0.9396 |
| 0.8046 | 3.3 | 2250 | 0.9385 |
| 1.5882 | 3.38 | 2300 | 0.9375 |
| 1.0068 | 3.45 | 2350 | 0.9366 |
| 1.205 | 3.52 | 2400 | 0.9357 |
| 0.689 | 3.6 | 2450 | 0.9350 |
| 0.8573 | 3.67 | 2500 | 0.9344 |
| 1.072 | 3.74 | 2550 | 0.9338 |
| 0.9188 | 3.82 | 2600 | 0.9332 |
| 1.3385 | 3.89 | 2650 | 0.9327 |
| 0.9067 | 3.96 | 2700 | 0.9324 |
| 0.9993 | 4.04 | 2750 | 0.9321 |
| 0.8222 | 4.11 | 2800 | 0.9317 |
| 0.8129 | 4.19 | 2850 | 0.9315 |
| 0.7861 | 4.26 | 2900 | 0.9313 |
| 1.3126 | 4.33 | 2950 | 0.9311 |
| 0.9465 | 4.41 | 3000 | 0.9310 |
| 0.9444 | 4.48 | 3050 | 0.9309 |
| 0.5677 | 4.55 | 3100 | 0.9308 |
| 0.7046 | 4.63 | 3150 | 0.9307 |
| 1.5036 | 4.7 | 3200 | 0.9307 |
| 1.0087 | 4.77 | 3250 | 0.9307 |
| 0.6705 | 4.85 | 3300 | 0.9306 |
| 1.0425 | 4.92 | 3350 | 0.9306 |
| 0.3666 | 4.99 | 3400 | 0.9306 |
Framework versions
- PEFT 0.10.1.dev0
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.15.2