metadata

license: other
base_model: google/gemma-2b
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: gemma_2b_twitter
    results: []

gemma_2b_twitter

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

Loss: 0.4890
Accuracy: 0.7702
F1 Macro: 0.7272
F1 Micro: 0.7702

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: 5e-06
train_batch_size: 16
eval_batch_size: 16
seed: 42
distributed_type: multi-GPU
num_devices: 2
total_train_batch_size: 32
total_eval_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro	F1 Micro
0.5354	0.18	50	0.5565	0.7307	0.6303	0.7307
0.5439	0.37	100	0.5230	0.7491	0.6894	0.7491
0.4586	0.55	150	0.5234	0.7390	0.6373	0.7390
0.4596	0.74	200	0.4985	0.7647	0.7211	0.7647
0.4886	0.92	250	0.4890	0.7702	0.7272	0.7702
0.3321	1.1	300	0.5282	0.7665	0.7135	0.7665
0.3257	1.29	350	0.5488	0.7583	0.7216	0.7583
0.2753	1.47	400	0.5489	0.7463	0.6899	0.7463
0.3038	1.65	450	0.5646	0.7610	0.7128	0.7610
0.3198	1.84	500	0.5672	0.7546	0.7165	0.7546
0.2154	2.02	550	0.5663	0.7619	0.7196	0.7619
0.1491	2.21	600	0.6345	0.7546	0.7079	0.7546
0.1073	2.39	650	0.7069	0.7564	0.7091	0.7564
0.1364	2.57	700	0.7230	0.7546	0.7133	0.7546
0.1545	2.76	750	0.7099	0.7518	0.7038	0.7518
0.1718	2.94	800	0.7138	0.7472	0.6925	0.7472

Framework versions

Transformers 4.39.0.dev0
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.15.2