See axolotl config

axolotl version: 0.4.0

adapter: qlora
base_model: mistralai/Mixtral-8x7B-Instruct-v0.1
bf16: true
chat_template: inst
dataset_prepared_path: last_run_prepared
datasets:
- conversation: mistral
  path: ./data/with_function_response/function_not_used_training_small.jsonl
  type: sharegpt
- conversation: mistral
  path: ./data/with_function_response/more_functions/function_used_training_small.jsonl
  type: sharegpt
debug: null
eval_max_new_tokens: 256
eval_steps: 0.2
eval_table_size: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: liuylhf/empower-functions-more-tools
learning_rate: 0.0002
load_in_4bit: true
load_in_8bit: false
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_model_dir: null
lora_r: 32
lora_target_modules:
- q_proj
- k_proj
- v_proj
- o_proj
loss_watchdog_patience: 3
loss_watchdog_threshold: 5.0
lr_scheduler: cosine
micro_batch_size: 2
model_config:
  output_router_logits: true
model_type: AutoModelForCausalLM
num_epochs: 2
optimizer: paged_adamw_8bit
output_dir: 2af0968cad514d6e9d5fb8448230e1c6/model
pad_to_sequence_len: true
sample_packing: true
save_steps: 0.1
sequence_len: 4096
strict: false
tf32: false
tokenizer_type: LlamaTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_log_model: end
wandb_name: mixtral-instruct-lora-no-negative
wandb_project: function-call
warmup_steps: 10
weight_decay: 0.0

empower-functions-more-tools

This model is a fine-tuned version of mistralai/Mixtral-8x7B-Instruct-v0.1 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0899

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: 0.0002
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• total_eval_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
1.8129	0.0	1	2.0522
0.1202	0.4	104	0.1054
0.1092	0.8	208	0.0976
0.0861	1.18	312	0.0938
0.0689	1.58	416	0.0908
0.0865	1.98	520	0.0899

Framework versions

• PEFT 0.9.0
• Transformers 4.38.2
• Pytorch 2.2.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.0