See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: Qwen/Qwen1.5-7B
bf16: auto
bnb_config_kwargs:
  bnb_4bit_quant_type: nf4
  bnb_4bit_use_double_quant: true
chat_template: llama3
cosine_min_lr_ratio: 0.1
data_processes: 16
dataset_prepared_path: null
datasets:
- data_files:
  - booksum_train_data.json
  ds_type: json
  path: /workspace/input_data/booksum_train_data.json
  type:
    field_input: book_id
    field_instruction: chapter
    field_output: summary
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map: '{'''':torch.cuda.current_device()}'
do_eval: true
early_stopping_patience: 1
eval_batch_size: 1
eval_sample_packing: false
eval_steps: 25
evaluation_strategy: steps
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 32
gradient_checkpointing: true
group_by_length: true
hub_model_id: cwaud/dd60ae62-f531-4ab6-9aea-f144b62d03ed
hub_repo: cwaud
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: 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
- v_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_memory:
  0: 70GiB
  1: 70GiB
  2: 70GiB
  3: 70GiB
max_steps: 190
micro_batch_size: 1
mlflow_experiment_name: /tmp/booksum_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
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: rayonlabs-rayon-labs
wandb_mode: online
wandb_project: Public_TuningSN
wandb_run: miner_id_24
wandb_runid: dd60ae62-f531-4ab6-9aea-f144b62d03ed
warmup_raio: 0.03
warmup_ratio: 0.04
weight_decay: 0.01
xformers_attention: null

dd60ae62-f531-4ab6-9aea-f144b62d03ed

This model is a fine-tuned version of Qwen/Qwen1.5-7B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.9462

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: 32
• total_train_batch_size: 128
• total_eval_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 7
• training_steps: 190

Training results

Training Loss	Epoch	Step	Validation Loss
2.3164	0.0103	1	2.2894
1.6573	0.2587	25	2.0379
1.6413	0.5175	50	1.9999
1.6448	0.7762	75	1.9803
2.149	1.0349	100	1.9675
2.1591	1.2937	125	1.9572
2.0887	1.5524	150	1.9498
2.1255	1.8111	175	1.9462

Framework versions

• PEFT 0.13.2
• Transformers 4.45.2
• Pytorch 2.4.1+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1