See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: Maykeye/TinyLLama-v0
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - fd22db0037d9e318_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/fd22db0037d9e318_train_data.json
  type:
    field_instruction: text_en
    field_output: text_cy
    format: '{instruction}'
    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: true
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: leixa/3380ee8a-44c1-452e-a34d-f9b4fc8a8138
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: 0
logging_steps: 3
lora_alpha: 128
lora_dropout: 0.1
lora_fan_in_fan_out: true
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_memory:
  0: 72GB
max_steps: 100
micro_batch_size: 8
mlflow_experiment_name: /tmp/fd22db0037d9e318_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: false
sample_packing: false
saves_per_epoch: 4
sequence_len: 1024
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: leixa-personal
wandb_mode: online
wandb_name: 3380ee8a-44c1-452e-a34d-f9b4fc8a8138
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 3380ee8a-44c1-452e-a34d-f9b4fc8a8138
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

3380ee8a-44c1-452e-a34d-f9b4fc8a8138

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: 7.7508

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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: 100

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0001	1	10.2267
9.9195	0.0007	9	9.6681
8.7825	0.0014	18	8.6971
8.3618	0.0021	27	8.2911
8.0958	0.0028	36	8.0980
7.9946	0.0034	45	7.9757
7.8897	0.0041	54	7.8835
7.81	0.0048	63	7.8218
7.822	0.0055	72	7.7827
7.6966	0.0062	81	7.7613
7.7321	0.0069	90	7.7524
7.6962	0.0076	99	7.7508

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1