This model is a fine-tuned version of deepseek-ai/deepseek-coder-1.3b-base on the generator dataset. It achieves the following results on the evaluation set:

Loss: 0.0031
Rewards/chosen: 0.3908
Rewards/rejected: -10.9590
Rewards/accuracies: 1.0
Rewards/margins: 11.3498
Logps/rejected: -238.4858
Logps/chosen: -106.5749
Logits/rejected: -19.3227
Logits/chosen: -18.9755

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: 1e-05
train_batch_size: 2
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 64
total_train_batch_size: 128
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 200
num_epochs: 2
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.1303	1.7067	100	0.0031	0.3908	-10.9590	1.0	11.3498	-238.4858	-106.5749	-19.3227	-18.9755

Framework versions

Transformers 4.43.0.dev0
Pytorch 2.2.2+cu121
Datasets 2.19.2
Tokenizers 0.19.1

stojchet/jndpo1

jndpo1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

stojchet/
jndpo1