metadata
base_model: Trisert/tinyllama-alpaca
library_name: peft
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: outputs/qlora-out-context
results: []
See axolotl config
axolotl version: 0.4.1
adapter: qlora
base_model: Trisert/tinyllama-alpaca
bf16: false
dataset_prepared_path: null
datasets:
- ds_tipe: json
path: /content/pubmed_continual_pretraning_dataset.jsonl
type: completion
debug: null
deepspeed: null
early_stopping_patience: null
eval_sample_packing: false
evals_per_epoch: 4
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
learning_rate: 0.0002
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 16
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lora_target_modules: null
lr_scheduler: cosine
micro_batch_size: 8
model_type: AutoModelForCausalLM
num_epochs: 4
optimizer: paged_adamw_32bit
output_dir: ./outputs/qlora-out-context
pad_to_sequence_len: false
resume_from_checkpoint: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 4096
special_tokens: null
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
val_set_size: 0.05
wandb_entity: null
wandb_log_model: null
wandb_name: null
wandb_project: null
wandb_watch: null
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null
outputs/qlora-out-context
This model is a fine-tuned version of Trisert/tinyllama-alpaca on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.8030
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- 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 |
|---|---|---|---|
| 2.6905 | 0.0336 | 1 | 2.7292 |
| 2.4725 | 0.2689 | 8 | 2.3972 |
| 1.9891 | 0.5378 | 16 | 2.0718 |
| 1.8345 | 0.8067 | 24 | 1.9329 |
| 1.8088 | 1.0756 | 32 | 1.8730 |
| 1.8183 | 1.3445 | 40 | 1.8430 |
| 1.8004 | 1.6134 | 48 | 1.8263 |
| 1.7674 | 1.8824 | 56 | 1.8167 |
| 1.7164 | 2.1513 | 64 | 1.8104 |
| 1.6525 | 2.4202 | 72 | 1.8069 |
| 1.7917 | 2.6891 | 80 | 1.8053 |
| 1.8022 | 2.9580 | 88 | 1.8037 |
| 1.6917 | 3.2269 | 96 | 1.8032 |
| 1.765 | 3.4958 | 104 | 1.8030 |
| 1.6784 | 3.7647 | 112 | 1.8030 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.1
- Pytorch 2.1.2+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1