Fischerboot/lil-biii-3B-adapter-lora

See axolotl config

axolotl version: 0.4.1

base_model: Fischerboot/Zephyr-3B-FreedomRP-Qlora-Merged
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: Fischerboot/mongotom-40k-alpaca
    type: alpaca
  - path: Fischerboot/freedom-rp-alpaca-shortend
    type: alpaca
  - path: Fischerboot/DAN-alpaca
    type: alpaca
dataset_prepared_path:
val_set_size: 0.05
output_dir: ./out/done

adapter: qlora
lora_model_dir:

sequence_len: 1024
sample_packing: true
pad_to_sequence_len: true

lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_modules:
lora_target_linear: true
lora_fan_in_fan_out:

wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 4
optimizer: paged_adamw_32bit
lr_scheduler: cosine
learning_rate: 0.0002

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

warmup_steps: 10
evals_per_epoch: 1
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:

out/done

This model is a fine-tuned version of Fischerboot/Zephyr-3B-FreedomRP-Qlora-Merged on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.0206

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
• gradient_accumulation_steps: 4
• total_train_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: 4

Training results

Training Loss	Epoch	Step	Validation Loss
2.9159	0.0029	1	2.9219
1.9943	0.9978	348	2.0447
2.0417	1.9849	696	1.9956
1.7099	2.9670	1044	2.0045
1.5156	3.9477	1392	2.0206

Framework versions

• PEFT 0.11.1
• Transformers 4.42.3
• Pytorch 2.1.2+cu118
• Datasets 2.19.1
• Tokenizers 0.19.1