metadata
language:
- en
license: apache-2.0
This is unsloth/llama-3-8b-Instruct trained on the Replete-AI/code-test-dataset using the code bellow with unsloth and google colab with under 15gb of vram. This training was complete in about 40 minutes total.
Copy from my announcement in my discord:
If anyone wants to train their own llama-3-8b model for free on any dataset that has around 1,500 lines of data or less you can now do it easily by using the code I provided in the model card for my test model in this repo and google colab. The training for this model uses (Unsloth + Qlora + Galore) to achieve the ability for training under such low vram.
For anyone that is new to coding and training Ai, all your really have to edit is
- (max_seq_length = 8192) To match the max tokens of the dataset or model you are using
- (model_name = "unsloth/llama-3-8b-Instruct",) Change what model you are finetuning, this setup is specifically for llama-3-8b
- (alpaca_prompt =) Change the prompt format, this one is setup to meet llama-3-8b-instruct format, but match it to your specifications.
- (dataset = load_dataset("Replete-AI/code-test-dataset", split = "train")) What dataset you are using from huggingface
- (model.push_to_hub_merged("rombodawg/test_dataset_Codellama-3-8B", tokenizer, save_method = "merged_16bit", token = ""))
- For the above you need to change "rombodawg" to your Hugginface name, "test_dataset_Codellama-3-8B" to the model name you want saved as, and in token = "" you need to put your huggingface write token so the model can be saved.
%%capture
import torch
major_version, minor_version = torch.cuda.get_device_capability()
# Must install separately since Colab has torch 2.2.1, which breaks packages
!pip install "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"
if major_version >= 8:
# Use this for new GPUs like Ampere, Hopper GPUs (RTX 30xx, RTX 40xx, A100, H100, L40)
!pip install --no-deps packaging ninja einops flash-attn xformers trl peft accelerate bitsandbytes
else:
# Use this for older GPUs (V100, Tesla T4, RTX 20xx)
!pip install --no-deps xformers trl peft accelerate bitsandbytes
pass
!pip install galore_torch
from unsloth import FastLanguageModel
import torch
max_seq_length = 8192 # Choose any! We auto support RoPE Scaling internally!
dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.
# 4bit pre quantized models we support for 4x faster downloading + no OOMs.
fourbit_models = [
"unsloth/mistral-7b-bnb-4bit",
"unsloth/mistral-7b-instruct-v0.2-bnb-4bit",
"unsloth/llama-2-7b-bnb-4bit",
"unsloth/gemma-7b-bnb-4bit",
"unsloth/gemma-7b-it-bnb-4bit", # Instruct version of Gemma 7b
"unsloth/gemma-2b-bnb-4bit",
"unsloth/gemma-2b-it-bnb-4bit", # Instruct version of Gemma 2b
"unsloth/llama-3-8b-bnb-4bit", # [NEW] 15 Trillion token Llama-3
] # More models at https://huggingface.co/unsloth
model, tokenizer = FastLanguageModel.from_pretrained(
model_name = "unsloth/llama-3-8b-Instruct",
max_seq_length = max_seq_length,
dtype = dtype,
load_in_4bit = load_in_4bit,
# token = "hf_...", # use one if using gated models like meta-llama/Llama-2-7b-hf
)
model = FastLanguageModel.get_peft_model(
model,
r = 16, # Choose any number > 0 ! Suggested 8, 16, 32, 64, 128
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
# [NEW] "unsloth" uses 30% less VRAM, fits 2x larger batch sizes!
use_gradient_checkpointing = "unsloth", # True or "unsloth" for very long context
random_state = 3407,
use_rslora = False, # We support rank stabilized LoRA
loftq_config = None, # And LoftQ
)
alpaca_prompt = """<|begin_of_text|><|start_header_id|>system<|end_header_id|>
Below is an instruction that describes a task, Write a response that appropriately completes the request.<|eot_id|><|start_header_id|>user<|end_header_id|>
{}<|eot_id|><|start_header_id|>assistant<|end_header_id|>{}"""
EOS_TOKEN = tokenizer.eos_token # Must add EOS_TOKEN
def formatting_prompts_func(examples):
inputs = examples["human"]
outputs = examples["assistant"]
texts = []
for input, output in zip(inputs, outputs):
# Must add EOS_TOKEN, otherwise your generation will go on forever!
text = alpaca_prompt.format(input, output) + EOS_TOKEN
texts.append(text)
return { "text" : texts, }
pass
from datasets import load_dataset
dataset = load_dataset("Replete-AI/code-test-dataset", split = "train")
dataset = dataset.map(formatting_prompts_func, batched = True,)
from trl import SFTTrainer
from transformers import TrainingArguments
from galore_torch import GaLoreAdamW8bit
import torch.nn as nn
galore_params = []
target_modules_list = ["attn", "mlp"]
for module_name, module in model.named_modules():
if not isinstance(module, nn.Linear):
continue
if not any(target_key in module_name for target_key in target_modules_list):
continue
print('mod ', module_name)
galore_params.append(module.weight)
id_galore_params = [id(p) for p in galore_params]
regular_params = [p for p in model.parameters() if id(p) not in id_galore_params]
param_groups = [{'params': regular_params},
{'params': galore_params, 'rank': 64, 'update_proj_gap': 200, 'scale': 0.25, 'proj_type': 'std'}]
optimizer = GaLoreAdamW8bit(param_groups, lr=2e-5)
trainer = SFTTrainer(
model = model,
tokenizer = tokenizer,
train_dataset = dataset,
optimizers=(optimizer, None),
dataset_text_field = "text",
max_seq_length = max_seq_length,
dataset_num_proc = 2,
packing = True, # Can make training 5x faster for short sequences.
args = TrainingArguments(
per_device_train_batch_size = 1,
gradient_accumulation_steps = 4,
warmup_steps = 5,
learning_rate = 2e-4,
fp16 = not torch.cuda.is_bf16_supported(),
bf16 = torch.cuda.is_bf16_supported(),
logging_steps = 1,
weight_decay = 0.01,
lr_scheduler_type = "linear",
seed = 3407,
output_dir = "outputs",
),
)
trainer_stats = trainer.train()
model.save_pretrained_merged("model", tokenizer, save_method = "merged_16bit",)
model.push_to_hub_merged("rombodawg/test_dataset_Codellama-3-8B", tokenizer, save_method = "merged_16bit", token = "")