Downloads · 30 days
9
26% of all-time downloads
llama_domar_pretuned
thorirhrafn/llama_domar_pretuned
llama_domar_pretuned 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.
should probably proofread and complete it, then remove this comment. --
All-time downloads
35
Public
Repo size
808 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors268 MB · 99%
At a glance
- Library
- peft
- Access
- Public
- Created
- Apr 11, 2024
- Updated
- Apr 14, 2024
- SHA
- 7b016e6f
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. -->
llama_domar_pretuned
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: 1.2652
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: 0.0001
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- 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 |
|---|---|---|---|
| 1.9648 | 0.02 | 500 | 1.9503 |
| 1.7555 | 0.05 | 1000 | 1.7822 |
| 1.7189 | 0.07 | 1500 | 1.6884 |
| 1.6753 | 0.1 | 2000 | 1.6275 |
| 1.5925 | 0.12 | 2500 | 1.5813 |
| 1.528 | 0.14 | 3000 | 1.5469 |
| 1.5106 | 0.17 | 3500 | 1.5173 |
| 1.5212 | 0.19 | 4000 | 1.4954 |
| 1.5137 | 0.22 | 4500 | 1.4752 |
| 1.4616 | 0.24 | 5000 | 1.4607 |
| 1.5058 | 0.26 | 5500 | 1.4453 |
| 1.4597 | 0.29 | 6000 | 1.4325 |
| 1.4399 | 0.31 | 6500 | 1.4211 |
| 1.4545 | 0.34 | 7000 | 1.4104 |
| 1.3372 | 0.36 | 7500 | 1.4015 |
| 1.3619 | 0.38 | 8000 | 1.3920 |
| 1.3949 | 0.41 | 8500 | 1.3848 |
| 1.3487 | 0.43 | 9000 | 1.3782 |
| 1.4019 | 0.45 | 9500 | 1.3716 |
| 1.4239 | 0.48 | 10000 | 1.3657 |
| 1.4097 | 0.5 | 10500 | 1.3612 |
| 1.3104 | 0.53 | 11000 | 1.3558 |
| 1.4048 | 0.55 | 11500 | 1.3494 |
| 1.4211 | 0.57 | 12000 | 1.3452 |
| 1.4037 | 0.6 | 12500 | 1.3420 |
| 1.2952 | 0.62 | 13000 | 1.3377 |
| 1.3429 | 0.65 | 13500 | 1.3338 |
| 1.3918 | 0.67 | 14000 | 1.3297 |
| 1.3201 | 0.69 | 14500 | 1.3269 |
| 1.3724 | 0.72 | 15000 | 1.3240 |
| 1.292 | 0.74 | 15500 | 1.3212 |
| 1.3172 | 0.77 | 16000 | 1.3185 |
| 1.3226 | 0.79 | 16500 | 1.3155 |
| 1.317 | 0.81 | 17000 | 1.3130 |
| 1.26 | 0.84 | 17500 | 1.3108 |
| 1.3303 | 0.86 | 18000 | 1.3084 |
| 1.3023 | 0.89 | 18500 | 1.3068 |
| 1.3527 | 0.91 | 19000 | 1.3051 |
| 1.2154 | 0.93 | 19500 | 1.3027 |
| 1.3466 | 0.96 | 20000 | 1.3008 |
| 1.3755 | 0.98 | 20500 | 1.2989 |
| 1.3004 | 1.01 | 21000 | 1.2976 |
| 1.3098 | 1.03 | 21500 | 1.2965 |
| 1.2149 | 1.05 | 22000 | 1.2957 |
| 1.3085 | 1.08 | 22500 | 1.2944 |
| 1.2719 | 1.1 | 23000 | 1.2927 |
| 1.2783 | 1.13 | 23500 | 1.2921 |
| 1.2772 | 1.15 | 24000 | 1.2907 |
| 1.2735 | 1.17 | 24500 | 1.2893 |
| 1.2668 | 1.2 | 25000 | 1.2887 |
| 1.3057 | 1.22 | 25500 | 1.2874 |
| 1.3161 | 1.25 | 26000 | 1.2865 |
| 1.3221 | 1.27 | 26500 | 1.2853 |
| 1.3028 | 1.29 | 27000 | 1.2843 |
| 1.2784 | 1.32 | 27500 | 1.2833 |
| 1.2999 | 1.34 | 28000 | 1.2831 |
| 1.2647 | 1.36 | 28500 | 1.2817 |
| 1.2258 | 1.39 | 29000 | 1.2812 |
| 1.2583 | 1.41 | 29500 | 1.2802 |
| 1.1942 | 1.44 | 30000 | 1.2794 |
| 1.2378 | 1.46 | 30500 | 1.2786 |
| 1.2946 | 1.48 | 31000 | 1.2781 |
| 1.1801 | 1.51 | 31500 | 1.2771 |
| 1.2644 | 1.53 | 32000 | 1.2765 |
| 1.1671 | 1.56 | 32500 | 1.2757 |
| 1.1899 | 1.58 | 33000 | 1.2752 |
| 1.2368 | 1.6 | 33500 | 1.2744 |
| 1.1986 | 1.63 | 34000 | 1.2738 |
| 1.2073 | 1.65 | 34500 | 1.2733 |
| 1.2625 | 1.68 | 35000 | 1.2727 |
| 1.225 | 1.7 | 35500 | 1.2721 |
| 1.2686 | 1.72 | 36000 | 1.2715 |
| 1.2642 | 1.75 | 36500 | 1.2713 |
| 1.2286 | 1.77 | 37000 | 1.2710 |
| 1.3091 | 1.8 | 37500 | 1.2704 |
| 1.3141 | 1.82 | 38000 | 1.2703 |
| 1.2973 | 1.84 | 38500 | 1.2699 |
| 1.2781 | 1.87 | 39000 | 1.2696 |
| 1.1794 | 1.89 | 39500 | 1.2693 |
| 1.261 | 1.92 | 40000 | 1.2689 |
| 1.2718 | 1.94 | 40500 | 1.2685 |
| 1.2768 | 1.96 | 41000 | 1.2682 |
| 1.2438 | 1.99 | 41500 | 1.2680 |
| 1.2777 | 2.01 | 42000 | 1.2679 |
| 1.2032 | 2.04 | 42500 | 1.2680 |
| 1.237 | 2.06 | 43000 | 1.2678 |
| 1.2169 | 2.08 | 43500 | 1.2676 |
| 1.2829 | 2.11 | 44000 | 1.2675 |
| 1.249 | 2.13 | 44500 | 1.2672 |
| 1.2304 | 2.16 | 45000 | 1.2672 |
| 1.2293 | 2.18 | 45500 | 1.2672 |
| 1.2781 | 2.2 | 46000 | 1.2670 |
| 1.2788 | 2.23 | 46500 | 1.2668 |
| 1.2381 | 2.25 | 47000 | 1.2666 |
| 1.1936 | 2.27 | 47500 | 1.2665 |
| 1.2306 | 2.3 | 48000 | 1.2662 |
| 1.2528 | 2.32 | 48500 | 1.2662 |
| 1.2482 | 2.35 | 49000 | 1.2661 |
| 1.2319 | 2.37 | 49500 | 1.2661 |
| 1.2272 | 2.39 | 50000 | 1.2660 |
| 1.2505 | 2.42 | 50500 | 1.2660 |
| 1.277 | 2.44 | 51000 | 1.2658 |
| 1.2377 | 2.47 | 51500 | 1.2658 |
| 1.2427 | 2.49 | 52000 | 1.2658 |
| 1.1909 | 2.51 | 52500 | 1.2657 |
| 1.2593 | 2.54 | 53000 | 1.2656 |
| 1.19 | 2.56 | 53500 | 1.2656 |
| 1.2662 | 2.59 | 54000 | 1.2654 |
| 1.2426 | 2.61 | 54500 | 1.2654 |
| 1.2474 | 2.63 | 55000 | 1.2654 |
| 1.2656 | 2.66 | 55500 | 1.2654 |
| 1.2121 | 2.68 | 56000 | 1.2653 |
| 1.2332 | 2.71 | 56500 | 1.2653 |
| 1.2549 | 2.73 | 57000 | 1.2653 |
| 1.2664 | 2.75 | 57500 | 1.2653 |
| 1.1769 | 2.78 | 58000 | 1.2653 |
| 1.2638 | 2.8 | 58500 | 1.2653 |
| 1.2316 | 2.83 | 59000 | 1.2652 |
| 1.2556 | 2.85 | 59500 | 1.2652 |
| 1.2116 | 2.87 | 60000 | 1.2652 |
| 1.287 | 2.9 | 60500 | 1.2652 |
| 1.2212 | 2.92 | 61000 | 1.2652 |
| 1.2871 | 2.95 | 61500 | 1.2652 |
| 1.2235 | 2.97 | 62000 | 1.2652 |
| 1.2176 | 2.99 | 62500 | 1.2652 |
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
- Created
- Apr 11, 2024
- Updated
- Apr 14, 2024