train_cornel_subj_cluster_hpoparams – AI Model by ErikBode | AlphaNeural AI

You can deploy this model and start earning money today!

ErikBode/
train_cornel_subj_cluster_hpoparams

like

0

generated_from_trainer

google/gemma-2b

Views

No views yet

Visualize in Weights & Biases

train_cornel_subj_cluster_hpoparams

This model is a fine-tuned version of google/gemma-2b on the None dataset. It achieves the following results on the evaluation set:

Loss: 3.7779
Rouge1: 0.3510
Rouge2: 0.0898
Rougel: 0.3129
Rougelsum: 0.3117

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: 0.00021
train_batch_size: 20
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 40
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 2
num_epochs: 2
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
3.8911	0.2041	25	4.0493	0.3395	0.0864	0.2962	0.2949
3.8512	0.4082	50	3.8527	0.3371	0.0826	0.2996	0.2986
3.7047	0.6122	75	3.8145	0.3388	0.0818	0.3022	0.3019
3.791	0.8163	100	3.7876	0.3496	0.0896	0.3057	0.3048
3.7131	1.0204	125	3.7734	0.3434	0.0850	0.3052	0.3045
3.4288	1.2245	150	3.7860	0.3562	0.0912	0.3147	0.3144
3.5349	1.4286	175	3.7748	0.3550	0.0859	0.3142	0.3136
3.3876	1.6327	200	3.7757	0.3516	0.0886	0.3121	0.3113
3.0759	1.8367	225	3.7779	0.3510	0.0898	0.3129	0.3117

Framework versions

PEFT 0.11.1
Transformers 4.42.4
Pytorch 2.3.1+cu121
Datasets 2.20.0
Tokenizers 0.19.1