Uploaded model
- Developed by: UsernameJustAnother
- License: apache-2.0
- Finetuned from model : unsloth/Mistral-Nemo-Instruct-2407
Experimental RP Finetune with secret sauce dataset, rsLoRA, r = 256, on an Colab A100 instance. 36GB vRAM used, 2 epochs ~ 3.5hrs of training.
This is for A/B testing vs Marlin v1, to see what difference rank 256 (v2) has compared to rank 64 (v1).
==((====))== Unsloth - 2x faster free finetuning | Num GPUs = 1
\\ /| Num examples = 8,160 | Num Epochs = 2
O^O/ \_/ \ Batch size per device = 2 | Gradient Accumulation steps = 4
\ / Total batch size = 8 | Total steps = 2,040
"-____-" Number of trainable parameters = 912,261,120
r = 256,
target_modules = ["q_proj", "k_proj", "v_proj", "o_proj",
"gate_proj", "up_proj", "down_proj",],
lora_alpha = 16,
lora_dropout = 0, # Supports any, but = 0 is optimized
bias = "none", # Supports any, but = "none" is optimized
use_gradient_checkpointing = "unsloth", # True or "unsloth" for very long context
random_state = 3407,
use_rslora = True, # lora_alpha --> 16
loftq_config = None,
per_device_train_batch_size = 2,
gradient_accumulation_steps = 4,
warmup_steps = 5,
num_train_epochs = 2,
learning_rate = 2e-5, # down from 2e-4, could go down to (5e-5 then 1e-5)
fp16 = not is_bfloat16_supported(),
bf16 = is_bfloat16_supported(),
logging_steps = 1,
optim = "adamw_8bit",
weight_decay = 0.01,
lr_scheduler_type = "linear",
seed = 3407,
This mistral model was trained 2x faster with Unsloth and Huggingface's TRL library.