See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: defog/llama-3-sqlcoder-8b
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 4ba7abd1a783cca6_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/4ba7abd1a783cca6_train_data.json
  type:
    field_input: system
    field_instruction: instruction
    field_output: chosen
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 256
eval_table_size: null
evals_per_epoch: 4
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 32
gradient_checkpointing: true
group_by_length: false
hub_model_id: tryingpro/02f47b36-6bb0-4259-99d9-53050fe333ac
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 3
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lora_target_modules:
- q_proj
- k_proj
- v_proj
- o_proj
- gate_proj
- down_proj
- up_proj
lr_scheduler: cosine
max_grad_norm: 2
max_steps: 90
micro_batch_size: 2
mlflow_experiment_name: /tmp/4ba7abd1a783cca6_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optim_args:
  adam_beta1: 0.9
  adam_beta2: 0.95
  adam_epsilon: 1.0e-05
optimizer: adamw_torch
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: 2048
special_tokens:
  pad_token: <|eot_id|>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: tryingpro-unicourt
wandb_mode: online
wandb_name: dde4cb64-07df-4e03-8a22-1f218483bad1
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: dde4cb64-07df-4e03-8a22-1f218483bad1
warmup_steps: 20
weight_decay: 0.02
xformers_attention: false

02f47b36-6bb0-4259-99d9-53050fe333ac

This model is a fine-tuned version of defog/llama-3-sqlcoder-8b on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7581

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 32
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-05
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 20
• training_steps: 90

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0004	1	1.1770
1.128	0.0032	8	1.0007
0.908	0.0064	16	0.8819
0.8148	0.0096	24	0.8319
0.8126	0.0128	32	0.8035
0.7618	0.0160	40	0.7878
0.8149	0.0192	48	0.7784
0.7839	0.0225	56	0.7703
0.7865	0.0257	64	0.7657
0.766	0.0289	72	0.7610
0.7596	0.0321	80	0.7588
0.7231	0.0353	88	0.7581

Framework versions

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