See axolotl config

axolotl version: 0.4.0

adapter: qlora
base_model: mistralai/Mixtral-8x22B-Instruct-v0.1
bf16: true
chat_template: inst
dataset_prepared_path: last_run_prepared
datasets:
- conversation: mistral
  path: ./data/with_function_response/original_clean/function_used_training.jsonl
  type: sharegpt
- conversation: mistral
  path: ./data/with_function_response/original_clean/function_not_used_training.jsonl
  type: sharegpt
- conversation: mistral
  path: ./data/with_function_response/parallel_call/parallel_data_training.jsonl
  type: sharegpt
debug: null
# eval_max_new_tokens: 256
# eval_steps: 0.2
# eval_table_size: null
flash_attention: true
fp16: false
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: liuylhf/parallel-call-original-4-epoch-mixtral-8x22b-instruct
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
lr_scheduler: cosine
micro_batch_size: 2
model_config:
  output_router_logits: true
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: paged_adamw_8bit
output_dir: model
pad_to_sequence_len: true
sample_packing: true
save_steps: 0.125
sequence_len: 4096
strict: false
tf32: false
tokenizer_type: LlamaTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0
wandb_log_model: end
wandb_name: more-tools
wandb_project: function-call
warmup_steps: 10
weight_decay: 0.0
fsdp:
  - full_shard
  - auto_wrap
fsdp_config:
  fsdp_limit_all_gathers: true
  fsdp_sync_module_states: true
  fsdp_offload_params: true
  fsdp_use_orig_params: false
  fsdp_cpu_ram_efficient_loading: true
  fsdp_transformer_layer_cls_to_wrap: MixtralSparseMoeBlock
  fsdp_state_dict_type: FULL_STATE_DICT
  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP

parallel-call-original-4-epoch-mixtral-8x22b-instruct

This model is a fine-tuned version of mistralai/Mixtral-8x22B-Instruct-v0.1 on an unknown dataset.

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: 4
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• total_eval_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 1

Framework versions

• PEFT 0.9.0
• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.0