See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: Intel/neural-chat-7b-v3-3
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - f341e63835691e5f_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/f341e63835691e5f_train_data.json
  type:
    field_instruction: question
    field_output: answer
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
do_eval: true
early_stopping_patience: 3
eval_batch_size: 4
eval_max_new_tokens: 128
eval_steps: 100
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: true
hub_model_id: lesso16/61e7ffc5-aec3-4bfa-8b8c-2ef75e886727
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.000216
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 50
lora_alpha: 128
lora_dropout: 0.15
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 700
micro_batch_size: 4
mlflow_experiment_name: /tmp/f341e63835691e5f_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 10
optimizer: adamw_torch_fused
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 100
saves_per_epoch: null
seed: 160
sequence_len: 1024
special_tokens:
  pad_token: </s>
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 3ba6d6aa-cbe1-4d32-97eb-906f9d7d26ad
wandb_project: 16a
wandb_run: your_name
wandb_runid: 3ba6d6aa-cbe1-4d32-97eb-906f9d7d26ad
warmup_steps: 100
weight_decay: 0.0
xformers_attention: null

61e7ffc5-aec3-4bfa-8b8c-2ef75e886727

This model is a fine-tuned version of Intel/neural-chat-7b-v3-3 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0527

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: 0.000216
• train_batch_size: 4
• eval_batch_size: 4
• seed: 160
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 100
• training_steps: 700

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0037	1	4.5437
2.4806	0.3661	100	0.4702
5.322	0.7323	200	0.6117
2.5778	1.0984	300	0.0979
2.5189	1.4645	400	0.0893
0.8023	1.8307	500	0.0894
0.401	2.1968	600	0.0727
0.3527	2.5629	700	0.0527

Framework versions

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