Edit model card

Transaminitis_L3_1000rate_1e6_SFT

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3409

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: 2
  • eval_batch_size: 1
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 4
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 100
  • training_steps: 1000

Training results

Training Loss Epoch Step Validation Loss
2.6005 0.2 25 2.3625
1.2395 0.4 50 0.8953
0.4393 0.6 75 0.4060
0.3066 0.8 100 0.3098
0.3 1.0 125 0.3041
0.2988 1.2 150 0.2955
0.2894 1.4 175 0.2894
0.2818 1.6 200 0.2810
0.278 1.8 225 0.2814
0.2716 2.0 250 0.2779
0.2648 2.2 275 0.2768
0.2628 2.4 300 0.2783
0.2624 2.6 325 0.2815
0.2635 2.8 350 0.2761
0.2556 3.0 375 0.2768
0.2408 3.2 400 0.2981
0.2309 3.4 425 0.2811
0.2461 3.6 450 0.2850
0.2332 3.8 475 0.2830
0.2428 4.0 500 0.2811
0.1987 4.2 525 0.3089
0.2113 4.4 550 0.3099
0.2108 4.6 575 0.3069
0.2068 4.8 600 0.3066
0.1927 5.0 625 0.3122
0.1758 5.2 650 0.3315
0.1749 5.4 675 0.3320
0.1751 5.6 700 0.3326
0.1744 5.8 725 0.3294
0.1698 6.0 750 0.3292
0.1621 6.2 775 0.3365
0.1532 6.4 800 0.3391
0.1638 6.6 825 0.3403
0.1587 6.8 850 0.3405
0.1641 7.0 875 0.3407
0.1659 7.2 900 0.3403
0.1567 7.4 925 0.3407
0.1626 7.6 950 0.3409
0.1544 7.8 975 0.3408
0.1611 8.0 1000 0.3409

Framework versions

  • Transformers 4.40.2
  • Pytorch 2.0.0+cu117
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
4
Safetensors
Model size
8.03B params
Tensor type
FP16
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for tsavage68/Transaminitis_L3_1000rate_1e6_SFT

Finetuned
(449)
this model