Modelpublic

tamang0000/tinyllama-assamese-v0.2.4

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes17downloads

Model Card

Before: কম্পিউটাৰ এন্ড এন্ডিং এন্ডিং এন্ডিং কম্পিউটাৰ এন্ড এন্ডিং এন্ডিং এন্ডিং

After: কম্পিউটাৰ অভিযান অভিযান অভিযান অভিযান কম্পিউটাৰ অভিযান অভিযান অভিযান অভিযান

max_seq_length = 512
per_device_train_batch_size = 2,
gradient_accumulation_steps = 4,
warmup_steps = 2,
max_steps = 10,
learning_rate = 0.0010,
fp16 = not torch.cuda.is_bf16_supported(),
bf16 = torch.cuda.is_bf16_supported(),
logging_steps = 1,
optim = "adamw_8bit",
weight_decay = 0.01,
lr_scheduler_type = "linear",
seed = 3407,
output_dir = "outputs",

Uploaded model

—Developed by: tamang0000
—License: apache-2.0
—Finetuned from model : unsloth/tinyllama-bnb-4bit

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.