zebra_4MLA12M2_8bt_SFT_8k_combined

This model is a fine-tuned version of meta-llama/Llama-3.2-1B-Instruct on the JunxiongWang/sftdatasetv3, the amd/OpenMathInstruct-2_ZebraLlama_2M, the amd/Zebra_Llama_OpenThoughts-114k-math, the amd/OpenR1-Math-220K and the amd/Zebra_Llama_ChatQA2-Long-SFT_long_sft_QA datasets. It achieves the following results on the evaluation set:

• Loss: 1436.5125

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: 6e-05
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 32
• total_eval_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
1522.4293	1.0	38873	1436.5125

Framework versions

• Transformers 4.43.1
• Pytorch 2.7.0a0+git6374332
• Datasets 2.20.0
• Tokenizers 0.19.1