model_phi_2_d_gender_bias_hate_bias_ep_1_7_a_sqn_a_b_p_100_5_v_16

This model is a fine-tuned version of microsoft/phi-2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7759
• Accuracy: 0.8339
• F1 Micro: 0.8339
• Auc: 0.9076

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: 16
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Micro	Auc
0.6088	1.0	38	0.6767	0.6719	0.6719	0.7535
0.4116	2.0	76	0.4517	0.7936	0.7936	0.8776
0.2824	3.0	114	0.5192	0.8100	0.8100	0.8880
0.0387	4.0	152	0.6496	0.8240	0.8240	0.8994
0.0353	5.0	190	0.7799	0.8248	0.8248	0.9007
0.0049	6.0	228	0.7434	0.8174	0.8174	0.9096
0.0036	7.0	266	0.7759	0.8339	0.8339	0.9076

Framework versions

• PEFT 0.11.1
• Transformers 4.44.0
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1