Hugging Face's logo

SethGA
/
neocortex

PEFT
Safetensors
llama
axolotl
Generated from Trainer
4-bit precision
bitsandbytes
Model card Files Community
neocortex / README.md
SethGA's picture
SethGA
Update README.md
3c0a76a verified
preview code
|
raw
history blame contribute delete
3.4 kB
metadata
library_name: peft
tags:
  - axolotl
  - generated_from_trainer
base_model: NousResearch/Llama-2-7b-hf
model-index:
  - name: neocortex
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.0

base_model: NousResearch/Llama-2-7b-hf
model_type: LlamaForCausalLM
tokenizer_type: LlamaTokenizer
is_llama_derived_model: true
hub_model_id: neocortex

load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: SethGA/neocortex
    type: alpaca
    shards: 20
dataset_prepared_path:
val_set_size: 0.05
output_dir: ./qlora-out

adapter: qlora
lora_model_dir:

sequence_len: 4096
sample_packing: false
eval_sample_packing: false
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: neocortex
wandb_entity:
wandb_watch:
wandb_run_id:
wandb_log_model: checkpoint

gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 3
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
eval_steps: 20
eval_table_size: 5
save_strategy: epoch
save_steps:
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
  bos_token: "<s>"
  eos_token: "</s>"
  unk_token: "<unk>"

neocortex

This model is a fine-tuned version of NousResearch/Llama-2-7b-hf on the neocortex_23k dataset. It achieves the following results on the evaluation set:

  • Loss: 1.4558

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: 3

Training results

Training Loss Epoch Step Validation Loss
1.5181 0.29 20 1.5627
1.437 0.58 40 1.4861
1.5196 0.87 60 1.4610
1.4037 1.16 80 1.4512
1.372 1.45 100 1.4493
1.3853 1.74 120 1.4424
1.2367 2.03 140 1.4460
1.283 2.32 160 1.4602
1.2933 2.61 180 1.4583
1.2397 2.9 200 1.4558

Framework versions

  • PEFT 0.8.2
  • Transformers 4.38.0.dev0
  • Pytorch 2.0.1+cu118
  • Datasets 2.17.0
  • Tokenizers 0.15.0