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:
  - 59b9daa5249384ab_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/59b9daa5249384ab_train_data.json
  type:
    field_instruction: context
    field_output: question
    format: '{instruction}'
    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: auxyus/e3659c20-bf7a-44f7-ba3e-e10e43cf241b
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/59b9daa5249384ab_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: 0a3400ba-5277-4ff0-9710-5a21e9e3a2a7
wandb_project: Gradients-On-Two
wandb_run: your_name
wandb_runid: 0a3400ba-5277-4ff0-9710-5a21e9e3a2a7
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

e3659c20-bf7a-44f7-ba3e-e10e43cf241b

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: 11.0391

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	0.0108	1	11.0752
44.2792	0.0968	9	11.0719
44.2529	0.1935	18	11.0627
44.2277	0.2903	27	11.0567
44.2022	0.3871	36	11.0524
44.1858	0.4839	45	11.0490
44.1543	0.5806	54	11.0463
44.1875	0.6774	63	11.0417
44.1534	0.7742	72	11.0402
44.1476	0.8710	81	11.0394
44.1507	0.9677	90	11.0390
44.146	1.0645	99	11.0391

Framework versions

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