Hugging Face's logo

shyp
/
saleperson_model

PEFT
Safetensors
trl
sft
Generated from Trainer
Model card Files Community
saleperson_model / README.md
shyp's picture
shyp
End of training
8cfacce verified
preview code
|
raw
history blame contribute delete
No virus
7.07 kB
metadata
license: llama3
library_name: peft
tags:
  - trl
  - sft
  - generated_from_trainer
base_model: meta-llama/Meta-Llama-3-8B
model-index:
  - name: saleperson_model
    results: []

saleperson_model

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

  • Loss: 0.3904

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: 0.0002
  • train_batch_size: 8
  • eval_batch_size: 16
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss
1.2687 0.09 10 1.0271
0.9208 0.18 20 0.8322
0.8525 0.27 30 0.7405
0.7371 0.36 40 0.6628
0.6385 0.44 50 0.6114
0.6129 0.53 60 0.5753
0.5744 0.62 70 0.5400
0.4751 0.71 80 0.5181
0.5396 0.8 90 0.5072
0.5591 0.89 100 0.4962
0.4722 0.98 110 0.4833
0.4386 1.07 120 0.4737
0.4628 1.16 130 0.4661
0.4762 1.24 140 0.4513
0.4393 1.33 150 0.4476
0.4075 1.42 160 0.4332
0.3876 1.51 170 0.4282
0.3927 1.6 180 0.4192
0.3692 1.69 190 0.4116
0.4409 1.78 200 0.4085
0.3214 1.87 210 0.3998
0.3653 1.96 220 0.4001
0.3436 2.04 230 0.4005
0.2995 2.13 240 0.3951
0.3033 2.22 250 0.3896
0.3767 2.31 260 0.3912
0.3236 2.4 270 0.3801
0.3008 2.49 280 0.3905
0.318 2.58 290 0.3710
0.3258 2.67 300 0.3707
0.2655 2.76 310 0.3742
0.3208 2.84 320 0.3719
0.3418 2.93 330 0.3667
0.2883 3.02 340 0.3729
0.2016 3.11 350 0.3692
0.2981 3.2 360 0.3704
0.2608 3.29 370 0.3762
0.29 3.38 380 0.3602
0.2496 3.47 390 0.3580
0.2669 3.56 400 0.3535
0.2206 3.64 410 0.3636
0.2624 3.73 420 0.3572
0.3052 3.82 430 0.3574
0.2789 3.91 440 0.3452
0.2465 4.0 450 0.3465
0.1996 4.09 460 0.3676
0.203 4.18 470 0.3540
0.2044 4.27 480 0.3692
0.2044 4.36 490 0.3497
0.2528 4.44 500 0.3585
0.2498 4.53 510 0.3505
0.1957 4.62 520 0.3549
0.2098 4.71 530 0.3495
0.2295 4.8 540 0.3478
0.2063 4.89 550 0.3409
0.2024 4.98 560 0.3478
0.185 5.07 570 0.3553
0.1571 5.16 580 0.3665
0.1969 5.24 590 0.3572
0.1719 5.33 600 0.3595
0.1874 5.42 610 0.3529
0.1953 5.51 620 0.3598
0.1545 5.6 630 0.3501
0.1947 5.69 640 0.3602
0.1804 5.78 650 0.3433
0.1498 5.87 660 0.3564
0.1722 5.96 670 0.3563
0.1607 6.04 680 0.3622
0.1728 6.13 690 0.3636
0.1464 6.22 700 0.3643
0.1514 6.31 710 0.3618
0.1522 6.4 720 0.3730
0.1378 6.49 730 0.3666
0.1363 6.58 740 0.3584
0.1312 6.67 750 0.3733
0.1304 6.76 760 0.3542
0.1399 6.84 770 0.3548
0.137 6.93 780 0.3645
0.139 7.02 790 0.3605
0.1167 7.11 800 0.3717
0.1205 7.2 810 0.3784
0.1107 7.29 820 0.3828
0.1258 7.38 830 0.3711
0.1186 7.47 840 0.3769
0.1094 7.56 850 0.3805
0.1405 7.64 860 0.3649
0.1331 7.73 870 0.3750
0.1199 7.82 880 0.3718
0.119 7.91 890 0.3669
0.1208 8.0 900 0.3707
0.1088 8.09 910 0.3801
0.1142 8.18 920 0.3848
0.1211 8.27 930 0.3772
0.1102 8.36 940 0.3781
0.11 8.44 950 0.3822
0.106 8.53 960 0.3846
0.0999 8.62 970 0.3869
0.0994 8.71 980 0.3859
0.0949 8.8 990 0.3843
0.1103 8.89 1000 0.3847
0.1032 8.98 1010 0.3850
0.1035 9.07 1020 0.3840
0.0941 9.16 1030 0.3878
0.1065 9.24 1040 0.3893
0.1011 9.33 1050 0.3909
0.1114 9.42 1060 0.3914
0.0961 9.51 1070 0.3908
0.0946 9.6 1080 0.3905
0.1044 9.69 1090 0.3904
0.1032 9.78 1100 0.3904
0.1009 9.87 1110 0.3904
0.0831 9.96 1120 0.3904

Framework versions

  • PEFT 0.7.1
  • Transformers 4.38.2
  • Pytorch 2.0.1+cu117
  • Datasets 2.14.4
  • Tokenizers 0.15.2