metadata
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
base_model: mistralai/Mixtral-8x7B-v0.1
model-index:
- name: qlora-out
results: []
See axolotl config
axolotl version: 0.4.0
base_model: mistralai/Mixtral-8x7B-v0.1
model_type: AutoModelForCausalLM
tokenizer_type: LlamaTokenizer
trust_remote_code: true
load_in_8bit: false
load_in_4bit: true
strict: false
datasets:
- path: ericflo/analysis-samples
type: sharegpt
ds_type: json
data_files:
- analysis-dataset-sharegpt-gpt4-sft-32x.jsonl
dataset_prepared_path: last_run_prepared
val_set_size: 0
output_dir: ./qlora-out
chat_template: chatml
model_config:
output_router_logits: true
adapter: qlora
lora_model_dir:
sequence_len: 32768
sample_packing: true
pad_to_sequence_len: true
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
wandb_project: analysismodel
wandb_entity:
wandb_watch:
wandb_name: am-mixtral-sft
wandb_log_model:
gradient_accumulation_steps: 8
micro_batch_size: 1
num_epochs: 6
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0005
train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false
gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true
loss_watchdog_threshold: 5.0
loss_watchdog_patience: 3
warmup_steps: 10
evals_per_epoch: 4
eval_table_size:
eval_max_new_tokens: 128
eval_sample_packing: false
saves_per_epoch: 0
save_strategy: "no"
debug:
weight_decay: 0.0
# fsdp:
# - full_shard
# fsdp_config:
# fsdp_transformer_layer_cls_to_wrap: MixtralSparseMoeBlock
deepspeed: deepspeed_configs/zero3_bf16.json
# 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_state_dict_type: SHARDED_STATE_DICT
special_tokens:
qlora-out
This model is a fine-tuned version of mistralai/Mixtral-8x7B-v0.1 on the None 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: 0.0005
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- 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: 6
Training results
Framework versions
- PEFT 0.10.0
- Transformers 4.40.0.dev0
- Pytorch 2.1.2+cu118
- Datasets 2.18.0
- Tokenizers 0.15.0