See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: tiiuae/falcon-7b
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 3e67dcd9c278ad31_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/3e67dcd9c278ad31_train_data.json
  type:
    field_instruction: Source
    field_output: Sentence
    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: 8
gradient_checkpointing: false
group_by_length: false
hub_model_id: sn56b2/c595e835-8e56-4c65-ac0b-fcd1e4c3d08b
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: 128
lora_dropout: 0.1
lora_fan_in_fan_out: true
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_steps: 500
micro_batch_size: 4
mlflow_experiment_name: /tmp/3e67dcd9c278ad31_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: false
sample_packing: false
saves_per_epoch: 4
sequence_len: 512
special_tokens:
  pad_token: <|endoftext|>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: sn56-miner
wandb_mode: disabled
wandb_name: null
wandb_project: god
wandb_run: pmq0
wandb_runid: null
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

c595e835-8e56-4c65-ac0b-fcd1e4c3d08b

This model is a fine-tuned version of tiiuae/falcon-7b on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.9978

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: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• total_eval_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: cosine
• lr_scheduler_warmup_steps: 10
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0037	1	5.8494
25.5609	0.1560	42	3.0784
23.2225	0.3120	84	2.8247
21.461	0.4680	126	2.6478
20.3326	0.6240	168	2.5137
18.8213	0.7799	210	2.3929
19.1582	0.9359	252	2.2803
14.7959	1.0919	294	2.2048
15.0965	1.2479	336	2.1123
14.1017	1.4039	378	2.0575
13.9084	1.5599	420	2.0130
13.3749	1.7159	462	1.9978

Framework versions

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