See axolotl config

axolotl version: 0.4.1

adapter: lora
auto_find_batch_size: true
base_model: sethuiyer/Medichat-Llama3-8B
bf16: auto
chat_template: llama3
dataloader_num_workers: 12
dataset_prepared_path: null
datasets:
- data_files:
  - a1b281c6d31a7336_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/a1b281c6d31a7336_train_data.json
  type:
    field_instruction: passage
    field_output: question
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 3
early_stopping_threshold: 0.001
eval_max_new_tokens: 128
eval_steps: 40
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: false
group_by_length: false
hub_model_id: mrferr3t/702c14dc-0985-4140-a663-6413d556bd0e
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0003
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 100
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: cosine
micro_batch_size: 32
mlflow_experiment_name: /tmp/a1b281c6d31a7336_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 50
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
s2_attention: null
sample_packing: false
save_steps: 40
saves_per_epoch: 0
sequence_len: 512
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.02
wandb_entity: null
wandb_mode: online
wandb_name: 12e67bb2-1212-4109-94de-02222dc25293
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 12e67bb2-1212-4109-94de-02222dc25293
warmup_ratio: 0.05
weight_decay: 0.0
xformers_attention: null

702c14dc-0985-4140-a663-6413d556bd0e

This model is a fine-tuned version of sethuiyer/Medichat-Llama3-8B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.0163

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: 0.0003
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Use adamw_bnb_8bit 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: 30
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0206	1	2.6607
No log	0.8247	40	1.4745
No log	1.6495	80	1.4676
1.2861	2.4742	120	1.5586
1.2861	3.2990	160	1.8056
0.3392	4.1237	200	2.0163

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.3.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.1