base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
load_in_8bit: false
load_in_4bit: true
strict: false
chat_template: llama3
datasets:
- path: yahma/alpaca-cleaned
  type: alpaca
  field_instruction: instruction
  field_input: input
  field_output: output
dataset_prepared_path: null
val_set_size: 0.0
output_dir: /workspace/axolotl/outputs
sequence_len: 256
sample_packing: false
pad_to_sequence_len: true
trust_remote_code: true
adapter: lora
lora_model_dir: null
lora_r: 8
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out: null
gradient_accumulation_steps: 1
micro_batch_size: 1
eval_batch_size: 1
num_epochs: 1
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002
train_on_inputs: false
group_by_length: false
bf16: auto
fp16: null
tf32: false
gradient_checkpointing: false
early_stopping_patience: null
resume_from_checkpoint: null
local_rank: null
logging_steps: 1
xformers_attention: null
flash_attention: true
s2_attention: null
wandb_project: Kurtos-Bittensor
wandb_entity: null
wandb_mode: offline
wandb_run: quick-test
wandb_runid: e2e-test-001
saves_per_epoch: 0
warmup_steps: 0
evals_per_epoch: 0
eval_table_size: null
eval_max_new_tokens: 128
max_steps: 3
debug: null
deepspeed: null
weight_decay: 0.0
fsdp: null
fsdp_config: null
wandb_name: e2e-test-001
hub_model_id: burckmeister/tinyllama-e2e-test
hub_strategy: end

tinyllama-e2e-test

This model is a fine-tuned version of TinyLlama/TinyLlama-1.1B-Chat-v1.0 on the yahma/alpaca-cleaned dataset.

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: 0.0002
train_batch_size: 1
eval_batch_size: 1
seed: 42
optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
training_steps: 3

Training results

Framework versions

PEFT 0.17.1
Transformers 4.57.1
Pytorch 2.7.1+cu126
Datasets 4.0.0
Tokenizers 0.22.1