See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: databricks/dolly-v2-3b
bf16: auto
chat_template: llama3
cosine_min_lr_ratio: 0.1
data_processes: 16
dataset_prepared_path: null
datasets:
- data_files:
  - 0c87c12065f9eb25_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/0c87c12065f9eb25_train_data.json
  type:
    field_instruction: human
    field_output: chosen
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map: '{'''':torch.cuda.current_device()}'
do_eval: true
early_stopping_patience: 20
eval_batch_size: 1
eval_sample_packing: false
eval_steps: 25
evaluation_strategy: steps
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 64
gradient_checkpointing: true
group_by_length: true
hub_model_id: sn56m4/a3a7ad8d-2c98-4f85-85db-608dff8d1d4f
hub_repo: stevemonite
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lora_target_modules:
- q_proj
- v_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_memory:
  0: 70GiB
max_steps: 277
micro_batch_size: 1
mlflow_experiment_name: /tmp/0c87c12065f9eb25_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 4
optim_args:
  adam_beta1: 0.9
  adam_beta2: 0.95
  adam_epsilon: 1e-5
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 50
save_strategy: steps
sequence_len: 2048
strict: false
tf32: false
tokenizer_type: AutoTokenizer
torch_compile: false
train_on_inputs: false
trust_remote_code: true
val_set_size: 50
wandb_entity: sn56-miner
wandb_mode: disabled
wandb_name: null
wandb_project: god
wandb_run: afe2
wandb_runid: null
warmup_raio: 0.03
warmup_ratio: 0.05
weight_decay: 0.01
xformers_attention: null

a3a7ad8d-2c98-4f85-85db-608dff8d1d4f

This model is a fine-tuned version of databricks/dolly-v2-3b on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.1880

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: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 64
• total_train_batch_size: 256
• total_eval_batch_size: 4
• 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-5
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 13
• training_steps: 277

Training results

Training Loss	Epoch	Step	Validation Loss
127.9479	0.0135	1	2.2311
46.5748	0.3386	25	1.2997
40.1518	0.6772	50	1.2226
94.6906	1.0233	75	1.2098
95.636	1.3619	100	1.2008
95.9536	1.7005	125	1.1938
87.5278	2.0466	150	1.1895
90.17	2.3852	175	1.1882
86.0495	2.7238	200	1.1881
81.9407	3.0698	225	1.1897
82.0073	3.4085	250	1.1889
78.4469	3.7471	275	1.1880

Framework versions

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