See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: fxmarty/really-tiny-falcon-testing
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 3915fdc6c710779f_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/3915fdc6c710779f_train_data.json
  type:
    field_input: input
    field_instruction: instruction
    field_output: output
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 4
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
gradient_clipping: 1.0
group_by_length: false
hub_model_id: Nexspear/8e9e3f5b-35ab-4385-b7ff-c31108cb7be3
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: 0
logging_steps: 3
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine
max_steps: 100
micro_batch_size: 8
mlflow_experiment_name: /tmp/3915fdc6c710779f_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
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: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: techspear-hub
wandb_mode: online
wandb_name: 13bb643e-4a1d-4cec-8e3e-fe2824e4af40
wandb_project: Gradients-On-Four
wandb_run: your_name
wandb_runid: 13bb643e-4a1d-4cec-8e3e-fe2824e4af40
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

8e9e3f5b-35ab-4385-b7ff-c31108cb7be3

This model is a fine-tuned version of fxmarty/really-tiny-falcon-testing on the None dataset. It achieves the following results on the evaluation set:

• Loss: 11.0468

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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: cosine
• lr_scheduler_warmup_steps: 10
• training_steps: 100

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0028	1	11.0862
44.3437	0.0251	9	11.0839
44.3211	0.0503	18	11.0785
44.2954	0.0754	27	11.0728
44.2735	0.1006	36	11.0670
44.247	0.1257	45	11.0610
44.2279	0.1508	54	11.0559
44.2096	0.1760	63	11.0516
44.1997	0.2011	72	11.0491
44.1925	0.2263	81	11.0477
44.185	0.2514	90	11.0470
44.1884	0.2765	99	11.0468

Framework versions

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