See axolotl config

axolotl version: 0.5.2

adapter: lora
base_model: Maykeye/TinyLLama-v0
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- format: custom
  path: mhenrichsen/alpaca_2k_test
  type:
    field_input: input
    field_instruction: instruction
    field_output: output
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 4
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: false
group_by_length: false
hub_model_id: samoline/e3699a47-8998-46cd-9792-b79e9d9eeb06
hub_repo: samoline
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 4
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 4
lora_target_linear: true
lr_scheduler: cosine
max_steps: 2
micro_batch_size: 1
mlflow_experiment_name: mhenrichsen/alpaca_2k_test
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 4
sequence_len: 512
special_tokens:
  pad_token: </s>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: samoline-nan
wandb_mode: online
wandb_name: e3947cb3-7c20-41c7-b215-9a5e96c2917b
wandb_project: Gradients-On-Demand
wandb_run: dev
wandb_runid: e3947cb3-7c20-41c7-b215-9a5e96c2917b
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

e3699a47-8998-46cd-9792-b79e9d9eeb06

This model is a fine-tuned version of Maykeye/TinyLLama-v0 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 7.2156

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
• lr_scheduler_warmup_steps: 10
• training_steps: 2

Training results

Training Loss	Epoch	Step	Validation Loss
7.6993	0.0005	1	7.2157
8.0312	0.0011	2	7.2156

Framework versions

• PEFT 0.13.2
• Transformers 4.46.3
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3