Downloads · 30 days
9
20% of all-time downloads
mistral-0.3b-backup
ayousanz/mistral-0.3b-backup
mistral-0.3b-backup is a text generation model from ayousanz. Use it when you need the model to write or continue text. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
46
Public
Parameters
355M
205 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors138 GB · 67%
At a glance
- Task
- Text Generation
- Library
- transformers
- Model type
- mistral
- Access
- Public
- Created
- Apr 10, 2024
- Updated
- Apr 17, 2024
- SHA
- 8c1ff9d5
Try a prompt
Base models
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
checkpoints-mistral-300M
This model is a fine-tuned version of None on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.4867
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: 6
- eval_batch_size: 6
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 16
- total_train_batch_size: 192
- total_eval_batch_size: 12
- optimizer: Adam with betas=(0.9,0.95) and epsilon=0.0001
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 4
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.5141 | 0.09 | 1000 | 4.5160 |
| 3.7879 | 0.18 | 2000 | 3.8531 |
| 3.5484 | 0.27 | 3000 | 3.5881 |
| 3.3734 | 0.36 | 4000 | 3.4287 |
| 3.2722 | 0.45 | 5000 | 3.3144 |
| 3.2276 | 0.54 | 6000 | 3.2299 |
| 3.1809 | 0.63 | 7000 | 3.1597 |
| 3.0706 | 0.72 | 8000 | 3.1043 |
| 3.0185 | 0.81 | 9000 | 3.0578 |
| 2.9496 | 0.9 | 10000 | 3.0157 |
| 2.9374 | 0.99 | 11000 | 2.9815 |
| 2.8794 | 1.08 | 12000 | 2.9487 |
| 2.8407 | 1.17 | 13000 | 2.9229 |
| 2.8818 | 1.26 | 14000 | 2.8973 |
| 2.8167 | 1.35 | 15000 | 2.8730 |
| 2.7941 | 1.44 | 16000 | 2.8515 |
| 2.7878 | 1.53 | 17000 | 2.8311 |
| 2.7894 | 1.62 | 18000 | 2.8113 |
| 2.7158 | 1.71 | 19000 | 2.7935 |
| 2.7409 | 1.8 | 20000 | 2.7765 |
| 2.7349 | 1.89 | 21000 | 2.7613 |
| 2.6631 | 1.98 | 22000 | 2.7451 |
| 2.6766 | 2.07 | 23000 | 2.7353 |
| 2.6405 | 2.16 | 24000 | 2.7231 |
| 2.6707 | 2.25 | 25000 | 2.7121 |
| 2.6362 | 2.34 | 26000 | 2.7005 |
| 2.5997 | 2.43 | 27000 | 2.6904 |
| 2.6549 | 2.52 | 28000 | 2.6798 |
| 2.6056 | 2.61 | 29000 | 2.6688 |
| 2.5722 | 2.7 | 30000 | 2.6594 |
| 2.6179 | 2.79 | 31000 | 2.6509 |
| 2.6064 | 2.88 | 32000 | 2.6423 |
| 2.5836 | 2.97 | 33000 | 2.6340 |
| 2.5502 | 3.06 | 34000 | 2.6285 |
| 2.5428 | 3.15 | 35000 | 2.6218 |
| 2.5342 | 3.24 | 36000 | 2.6160 |
| 2.5152 | 3.33 | 37000 | 2.6090 |
| 2.5138 | 5.13 | 38000 | 2.5766 |
| 2.5032 | 5.27 | 39000 | 2.5683 |
| 2.4783 | 5.4 | 40000 | 2.5609 |
| 2.4519 | 5.54 | 41000 | 2.5545 |
| 2.4918 | 5.67 | 42000 | 2.5472 |
| 2.4591 | 5.81 | 43000 | 2.5411 |
| 2.4756 | 5.94 | 44000 | 2.5354 |
| 2.4434 | 6.08 | 45000 | 2.5345 |
| 2.4312 | 6.21 | 46000 | 2.5301 |
| 2.4576 | 6.35 | 47000 | 2.5242 |
| 2.4343 | 6.48 | 48000 | 2.5192 |
| 2.426 | 6.62 | 49000 | 2.5139 |
| 2.4136 | 6.75 | 50000 | 2.5084 |
| 2.4463 | 6.89 | 51000 | 2.5037 |
| 2.345 | 7.02 | 52000 | 2.5016 |
| 2.3736 | 7.16 | 53000 | 2.4990 |
| 2.4092 | 7.29 | 54000 | 2.4955 |
| 2.3689 | 7.43 | 55000 | 2.4917 |
| 2.3797 | 7.56 | 56000 | 2.4867 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.2+cu121
- Datasets 2.14.5
- Tokenizers 0.14.1
- Task
- Text Generation
- Library
- transformers
- Type
- mistral
- Created
- Apr 10, 2024
- Updated
- Apr 17, 2024