metadata

license: mit
library_name: peft
tags:
  - trl
  - sft
  - generated_from_trainer
base_model: microsoft/Phi-3-mini-128k-instruct
datasets:
  - generator
model-index:
  - name: Phi-3-mini-128k-instruct-advisegpt-v0.3
    results: []

Phi-3-mini-128k-instruct-advisegpt-v0.3

This model is a fine-tuned version of microsoft/Phi-3-mini-128k-instruct on the generator dataset. It achieves the following results on the evaluation set:

Loss: 1.6513

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: 2e-05
train_batch_size: 14
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 30
total_train_batch_size: 420
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
num_epochs: 3
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.838	0.9633	14	1.7877
1.5618	1.9954	29	1.6674
1.4944	2.8899	42	1.6513

Framework versions

PEFT 0.10.0
Transformers 4.40.2
Pytorch 2.2.0
Datasets 2.19.1
Tokenizers 0.19.1