Downloads · 30 days
6
17% of all-time downloads
icellama_domar_pretuned_v2
thorirhrafn/icellama_domar_pretuned_v2
icellama_domar_pretuned_v2 is a machine learning model from thorirhrafn. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as llama2.
should probably proofread and complete it, then remove this comment. --
All-time downloads
35
Public
Repo size
6.4 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors2.1 GB · 100%
At a glance
- Library
- peft
- License
- llama2
- Access
- Public
- Created
- Apr 16, 2024
- Updated
- Apr 18, 2024
- SHA
- 4c4a0c50
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. -->
icellama_domar_pretuned_v2
This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.5311
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-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.6301 | 0.05 | 500 | 1.6427 |
| 1.6438 | 0.1 | 1000 | 1.6136 |
| 1.612 | 0.14 | 1500 | 1.5963 |
| 1.5958 | 0.19 | 2000 | 1.5836 |
| 1.5981 | 0.24 | 2500 | 1.5741 |
| 1.61 | 0.29 | 3000 | 1.5666 |
| 1.5862 | 0.34 | 3500 | 1.5608 |
| 1.529 | 0.38 | 4000 | 1.5563 |
| 1.5807 | 0.43 | 4500 | 1.5523 |
| 1.5732 | 0.48 | 5000 | 1.5491 |
| 1.553 | 0.53 | 5500 | 1.5464 |
| 1.5993 | 0.57 | 6000 | 1.5444 |
| 1.5427 | 0.62 | 6500 | 1.5424 |
| 1.5705 | 0.67 | 7000 | 1.5408 |
| 1.546 | 0.72 | 7500 | 1.5395 |
| 1.5425 | 0.77 | 8000 | 1.5383 |
| 1.5793 | 0.81 | 8500 | 1.5374 |
| 1.5514 | 0.86 | 9000 | 1.5366 |
| 1.5709 | 0.91 | 9500 | 1.5359 |
| 1.5638 | 0.96 | 10000 | 1.5352 |
| 1.5792 | 1.01 | 10500 | 1.5347 |
| 1.5027 | 1.05 | 11000 | 1.5342 |
| 1.5345 | 1.1 | 11500 | 1.5339 |
| 1.5431 | 1.15 | 12000 | 1.5334 |
| 1.548 | 1.2 | 12500 | 1.5332 |
| 1.5513 | 1.25 | 13000 | 1.5328 |
| 1.5717 | 1.29 | 13500 | 1.5326 |
| 1.539 | 1.34 | 14000 | 1.5324 |
| 1.5256 | 1.39 | 14500 | 1.5323 |
| 1.4934 | 1.44 | 15000 | 1.5320 |
| 1.5648 | 1.48 | 15500 | 1.5320 |
| 1.5342 | 1.53 | 16000 | 1.5319 |
| 1.5125 | 1.58 | 16500 | 1.5317 |
| 1.4974 | 1.63 | 17000 | 1.5316 |
| 1.5086 | 1.68 | 17500 | 1.5316 |
| 1.5387 | 1.72 | 18000 | 1.5315 |
| 1.5126 | 1.77 | 18500 | 1.5314 |
| 1.565 | 1.82 | 19000 | 1.5314 |
| 1.5888 | 1.87 | 19500 | 1.5313 |
| 1.5585 | 1.92 | 20000 | 1.5313 |
| 1.5599 | 1.96 | 20500 | 1.5313 |
| 1.5476 | 2.01 | 21000 | 1.5312 |
| 1.5661 | 2.06 | 21500 | 1.5312 |
| 1.5514 | 2.11 | 22000 | 1.5312 |
| 1.531 | 2.16 | 22500 | 1.5312 |
| 1.5705 | 2.2 | 23000 | 1.5311 |
| 1.5576 | 2.25 | 23500 | 1.5311 |
| 1.5601 | 2.3 | 24000 | 1.5312 |
| 1.5438 | 2.35 | 24500 | 1.5312 |
| 1.5758 | 2.39 | 25000 | 1.5311 |
| 1.5704 | 2.44 | 25500 | 1.5311 |
| 1.5225 | 2.49 | 26000 | 1.5311 |
| 1.556 | 2.54 | 26500 | 1.5311 |
| 1.5701 | 2.59 | 27000 | 1.5311 |
| 1.5135 | 2.63 | 27500 | 1.5311 |
| 1.5029 | 2.68 | 28000 | 1.5311 |
| 1.5471 | 2.73 | 28500 | 1.5311 |
| 1.5083 | 2.78 | 29000 | 1.5311 |
| 1.5557 | 2.83 | 29500 | 1.5311 |
| 1.5394 | 2.87 | 30000 | 1.5311 |
| 1.5449 | 2.92 | 30500 | 1.5311 |
| 1.5467 | 2.97 | 31000 | 1.5311 |
Framework versions
- PEFT 0.8.2
- Transformers 4.38.1
- Pytorch 2.2.0+cu118
- Datasets 2.17.1
- Tokenizers 0.15.2
- Library
- peft
- License
- llama2
- Created
- Apr 16, 2024
- Updated
- Apr 18, 2024