README.md

metadata

library_name: peft
base_model: katuni4ka/tiny-random-qwen1.5-moe
tags:
  - axolotl
  - generated_from_trainer
model-index:
  - name: 5663e6cc-de8d-4ea7-87d4-be498fce4d0e
    results: []

5663e6cc-de8d-4ea7-87d4-be498fce4d0e

This model is a fine-tuned version of katuni4ka/tiny-random-qwen1.5-moe on the None dataset. It achieves the following results on the evaluation set:

• Loss: 11.7402

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: 0.000206
• train_batch_size: 4
• eval_batch_size: 4
• seed: 60
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• training_steps: 9000

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0003	1	11.9364
11.7934	0.1516	500	11.7890
11.7818	0.3033	1000	11.7749
11.7769	0.4549	1500	11.7676
11.772	0.6066	2000	11.7620
11.7681	0.7582	2500	11.7579
11.76	0.9098	3000	11.7542
11.7532	1.0615	3500	11.7502
11.774	1.2131	4000	11.7484
11.7583	1.3648	4500	11.7457
11.7423	1.5164	5000	11.7446
11.7536	1.6681	5500	11.7429
11.7444	1.8197	6000	11.7420
11.753	1.9713	6500	11.7410
11.7538	2.1230	7000	11.7406
11.7483	2.2746	7500	11.7404
11.7539	2.4263	8000	11.7401
11.7568	2.5779	8500	11.7402
11.7279	2.7295	9000	11.7402

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1