mlfoundations-dev / simpo-oh_teknium_scaling_down_random_0.4

huggingface.co
Total runs: 15
24-hour runs: 0
7-day runs: 1
30-day runs: 15
Model's Last Updated: January 17 2025
text-generation

Introduction of simpo-oh_teknium_scaling_down_random_0.4

Model Details of simpo-oh_teknium_scaling_down_random_0.4

simpo-oh_teknium_scaling_down_random_0.4

This model is a fine-tuned version of mlfoundations-dev/oh_teknium_scaling_down_random_0.4 on the mlfoundations-dev/gemma2-ultrafeedback-armorm dataset. It achieves the following results on the evaluation set:

  • Loss: 2.9306
  • Rewards/chosen: -28.9839
  • Rewards/rejected: -34.2554
  • Rewards/accuracies: 0.7574
  • Rewards/margins: 5.2715
  • Logps/chosen: -2.8984
  • Logps/rejected: -3.4255
  • Logits/chosen: -1.0814
  • Logits/rejected: -1.0757
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: 8e-07
  • train_batch_size: 2
  • eval_batch_size: 2
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 128
  • total_eval_batch_size: 16
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 1.0
Training results
Training Loss Epoch Step Validation Loss Rewards/chosen Rewards/rejected Rewards/accuracies Rewards/margins Logps/chosen Logps/rejected Logits/chosen Logits/rejected
2.6473 0.9997 442 2.9306 -28.9839 -34.2554 0.7574 5.2715 -2.8984 -3.4255 -1.0814 -1.0757
Framework versions
  • Transformers 4.46.1
  • Pytorch 2.3.0
  • Datasets 3.1.0
  • Tokenizers 0.20.3

Runs of mlfoundations-dev simpo-oh_teknium_scaling_down_random_0.4 on huggingface.co

15
Total runs
0
24-hour runs
0
3-day runs
1
7-day runs
15
30-day runs

More Information About simpo-oh_teknium_scaling_down_random_0.4 huggingface.co Model

More simpo-oh_teknium_scaling_down_random_0.4 license Visit here:

https://choosealicense.com/licenses/llama3.1

simpo-oh_teknium_scaling_down_random_0.4 huggingface.co

simpo-oh_teknium_scaling_down_random_0.4 huggingface.co is an AI model on huggingface.co that provides simpo-oh_teknium_scaling_down_random_0.4's model effect (), which can be used instantly with this mlfoundations-dev simpo-oh_teknium_scaling_down_random_0.4 model. huggingface.co supports a free trial of the simpo-oh_teknium_scaling_down_random_0.4 model, and also provides paid use of the simpo-oh_teknium_scaling_down_random_0.4. Support call simpo-oh_teknium_scaling_down_random_0.4 model through api, including Node.js, Python, http.

simpo-oh_teknium_scaling_down_random_0.4 huggingface.co Url

https://huggingface.co/mlfoundations-dev/simpo-oh_teknium_scaling_down_random_0.4

mlfoundations-dev simpo-oh_teknium_scaling_down_random_0.4 online free

simpo-oh_teknium_scaling_down_random_0.4 huggingface.co is an online trial and call api platform, which integrates simpo-oh_teknium_scaling_down_random_0.4's modeling effects, including api services, and provides a free online trial of simpo-oh_teknium_scaling_down_random_0.4, you can try simpo-oh_teknium_scaling_down_random_0.4 online for free by clicking the link below.

mlfoundations-dev simpo-oh_teknium_scaling_down_random_0.4 online free url in huggingface.co:

https://huggingface.co/mlfoundations-dev/simpo-oh_teknium_scaling_down_random_0.4

simpo-oh_teknium_scaling_down_random_0.4 install

simpo-oh_teknium_scaling_down_random_0.4 is an open source model from GitHub that offers a free installation service, and any user can find simpo-oh_teknium_scaling_down_random_0.4 on GitHub to install. At the same time, huggingface.co provides the effect of simpo-oh_teknium_scaling_down_random_0.4 install, users can directly use simpo-oh_teknium_scaling_down_random_0.4 installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

simpo-oh_teknium_scaling_down_random_0.4 install url in huggingface.co:

https://huggingface.co/mlfoundations-dev/simpo-oh_teknium_scaling_down_random_0.4

Url of simpo-oh_teknium_scaling_down_random_0.4

simpo-oh_teknium_scaling_down_random_0.4 huggingface.co Url

Provider of simpo-oh_teknium_scaling_down_random_0.4 huggingface.co

mlfoundations-dev
ORGANIZATIONS

Other API from mlfoundations-dev