See axolotl config

axolotl version: 0.4.1

adapter: qlora
auto_resume_from_checkpoints: false
base_model: scb10x/llama-3-typhoon-v1.5-8b-instruct
bf16: false
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - f61bec5b82911b5f_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/f61bec5b82911b5f_train_data.json
  type:
    field_input: genre
    field_instruction: premise
    field_output: hypothesis
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 4
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: true
fp16: true
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 32
gradient_checkpointing: true
group_by_length: false
hub_model_id: error577/52d32e8c-e6c5-404b-8326-b6845f7443fd
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 128
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: null
micro_batch_size: 1
mlflow_experiment_name: /tmp/f61bec5b82911b5f_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_torch_4bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 100
sequence_len: 128
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.0005
wandb_entity: null
wandb_mode: online
wandb_name: 0902982d-d12a-4780-a62c-f6980404ea63
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 0902982d-d12a-4780-a62c-f6980404ea63
warmup_steps: 30
weight_decay: 0.0
xformers_attention: null

52d32e8c-e6c5-404b-8326-b6845f7443fd

This model is a fine-tuned version of scb10x/llama-3-typhoon-v1.5-8b-instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.8018

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
• gradient_accumulation_steps: 32
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_4BIT 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: 30
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
3.5146	0.0001	1	3.9565
1.8081	0.0078	100	1.7830
1.8802	0.0155	200	1.7761
1.585	0.0233	300	1.8033
1.6037	0.0311	400	1.7851
1.5661	0.0388	500	1.8069
1.8584	0.0466	600	1.8018

Framework versions

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