You can deploy this model and start earning money today!

anakiou/
mellum2-12b-instruct-modality-conflict-qlora

Name: mellum2-12b-instruct-modality-conflict-qlora
Availability: InStock

JetBrains/Mellum2-12B-A2.5B-Instruct

apache-2.0

Views

No views yet

mellum2-12b-instruct-modality-conflict-qlora

This model is a fine-tuned version of JetBrains/Mellum2-12B-A2.5B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.2634

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: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 16
optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT 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: 0.05
num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
0.3835	0.3161	50	0.3682
0.3504	0.6322	100	0.3083
0.3326	0.9482	150	0.2948
0.2676	1.2592	200	0.2763
0.2362	1.5753	250	0.2679
0.2301	1.8913	300	0.2630
0.2834	2.0	318	0.2634

Framework versions

PEFT 0.19.1
Transformers 5.12.1
Pytorch 2.11.0+cu128
Datasets 5.0.0
Tokenizers 0.22.2

anakiou/mellum2-12b-instruct-modality-conflict-qlora

mellum2-12b-instruct-modality-conflict-qlora

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

anakiou/
mellum2-12b-instruct-modality-conflict-qlora