adpretko/AnghaBench-armv8-O2-native-clang-full-coder-epoch1-AMD

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

AnghaBench-armv8-O2-native-clang-full-coder-epoch1-AMD

This model is a fine-tuned version of Qwen/Qwen2.5-Coder-1.5B on the AnghaBench-armv8-O2-native-clang-fullpart00, the AnghaBench-armv8-O2-native-clang-fullpart01, the AnghaBench-armv8-O2-native-clang-fullpart02, the AnghaBench-armv8-O2-native-clang-fullpart03, the AnghaBench-armv8-O2-native-clang-fullpart04, the AnghaBench-armv8-O2-native-clang-fullpart05, the AnghaBench-armv8-O2-native-clang-fullpart06, the AnghaBench-armv8-O2-native-clang-fullpart07, the AnghaBench-armv8-O2-native-clang-fullpart08, the AnghaBench-armv8-O2-native-clang-fullpart09, the AnghaBench-armv8-O2-native-clang-fullpart10, the AnghaBench-armv8-O2-native-clang-fullpart11, the AnghaBench-armv8-O2-native-clang-fullpart12, the AnghaBench-armv8-O2-native-clang-fullpart13, the AnghaBench-armv8-O2-native-clang-fullpart14, the AnghaBench-armv8-O2-native-clang-fullpart15, the AnghaBench-armv8-O2-native-clang-fullpart16, the AnghaBench-armv8-O2-native-clang-fullpart17, the AnghaBench-armv8-O2-native-clang-fullpart18, the AnghaBench-armv8-O2-native-clang-fullpart19, the AnghaBench-armv8-O2-native-clang-fullpart20, the AnghaBench-armv8-O2-native-clang-fullpart21, the AnghaBench-armv8-O2-native-clang-fullpart22, the AnghaBench-armv8-O2-native-clang-fullpart23, the AnghaBench-armv8-O2-native-clang-fullpart24, the AnghaBench-armv8-O2-native-clang-fullpart25, the AnghaBench-armv8-O2-native-clang-fullpart26, the AnghaBench-armv8-O2-native-clang-fullpart27, the AnghaBench-armv8-O2-native-clang-fullpart28, the AnghaBench-armv8-O2-native-clang-fullpart29, the AnghaBench-armv8-O2-native-clang-fullpart30, the AnghaBench-armv8-O2-native-clang-fullpart31, the AnghaBench-armv8-O2-native-clang-fullpart32, the AnghaBench-armv8-O2-native-clang-fullpart33, the AnghaBench-armv8-O2-native-clang-fullpart34, the AnghaBench-armv8-O2-native-clang-fullpart35, the AnghaBench-armv8-O2-native-clang-fullpart36, the AnghaBench-armv8-O2-native-clang-fullpart37, the AnghaBench-armv8-O2-native-clang-fullpart38, the AnghaBench-armv8-O2-native-clang-fullpart39, the AnghaBench-armv8-O2-native-clang-fullpart40, the AnghaBench-armv8-O2-native-clang-fullpart41, the AnghaBench-armv8-O2-native-clang-fullpart42, the AnghaBench-armv8-O2-native-clang-fullpart43, the AnghaBench-armv8-O2-native-clang-fullpart44, the AnghaBench-armv8-O2-native-clang-fullpart45, the AnghaBench-armv8-O2-native-clang-fullpart46, the AnghaBench-armv8-O2-native-clang-fullpart47, the AnghaBench-armv8-O2-native-clang-fullpart48 and the AnghaBench-armv8-O2-native-clang-fullpart49 datasets.

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: 2e-05
• —trainbatchsize: 8
• —evalbatchsize: 8
• —seed: 42
• —distributed_type: multi-GPU
• —num_devices: 8
• —gradientaccumulationsteps: 8
• —totaltrainbatch_size: 512
• —totalevalbatch_size: 64
• —optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• —lrschedulertype: cosine
• —lrschedulerwarmup_ratio: 0.1
• —num_epochs: 1.0

Training results

Framework versions

• —Transformers 4.55.0
• —Pytorch 2.8.0+rocm6.3
• —Datasets 3.6.0
• —Tokenizers 0.21.1