See axolotl config

axolotl version: 0.10.0.dev0

adapter: lora
base_model: bigscience/bloomz-560m
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 6f6301de67b5869a_train_data.json
  ds_type: json
  path: /workspace/input_data/
  split: train
  type: completion
  field: prompt
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_steps: 10
eval_table_size: null
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hf_use_auth_token: true
hub_model_id: eeeebbb2/4686d268-dae0-47a7-9278-3db4f7d4d27d
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.00002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 16
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 8
lora_target_linear: true
lr_scheduler: constant
max_steps: 10
micro_batch_size: 2
max_grad_norm: 0.5
mlflow_experiment_name: /tmp/6f6301de67b5869a_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1                     
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
push_dataset_to_hub: "false"
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 10
save_strategy: steps
save_total_limit: 1
sequence_len: 512
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 17436655-754e-42da-9e74-3a4a060545c7
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 17436655-754e-42da-9e74-3a4a060545c7
warmup_steps: 0
weight_decay: 0.0
xformers_attention: null

4686d268-dae0-47a7-9278-3db4f7d4d27d

This model is a fine-tuned version of bigscience/bloomz-560m on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 4.6224

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: 2e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• 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: constant
• training_steps: 10

Training results

Training Loss	Epoch	Step	Validation Loss
2.8032	0.0006	1	4.6840
3.1281	0.0061	10	4.6224

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.5.1+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1