base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
model_type: LlamaForCausalLM
tokenizer_type: LlamaTokenizer

load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: imdb_1k_alpaca.jsonl
    type: alpaca
dataset_prepared_path: last_run_prepared
val_set_size: 0.05
output_dir: ./outputs/qlora-out
hub_model_id: satish860/hc-tinyllama-alpaca

adapter: qlora
lora_model_dir:

sequence_len: 4096
sample_packing: true
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:
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: 4
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:

hc-tinyllama-alpaca

This model is a fine-tuned version of TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0225

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
8.0151	0.0851	1	7.9045
7.967	0.2553	3	7.7350
6.5633	0.5106	6	4.8012
1.5838	0.7660	9	0.7756
0.2319	1.0213	12	0.1592
0.0669	1.2340	15	0.0973
0.0344	1.4894	18	0.0453
0.1146	1.7447	21	0.0754
0.0896	2.0	24	0.0517
0.0293	2.2340	27	0.0486
0.0378	2.4894	30	0.0566
0.0523	2.7447	33	0.0270
0.0886	3.0	36	0.0226
0.0504	3.2128	39	0.0232
0.089	3.4681	42	0.0225

Framework versions

PEFT 0.10.0
Transformers 4.40.2
Pytorch 2.1.2+cu118
Datasets 2.19.1
Tokenizers 0.19.1