|
--- |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: out |
|
results: [] |
|
--- |
|
### This is the Instruction Fine Tuned version of [Tiny Llama](https://github.com/jzhang38/TinyLlama) on [@Teknium1's](https://twitter.com/Teknium1) [openhermes](https://huggingface.co/datasets/teknium/openhermes) dataset. |
|
|
|
`"The TinyLlama project aims to pretrain a 1.1B Llama model on 3 trillion tokens. With some proper optimization, we can achieve this within a span of "just" 90 days using 16 A100-40G GPUs 🚀🚀. The training has started on 2023-09-01."` |
|
|
|
<details><summary>See axolotl config</summary> |
|
|
|
axolotl version: `0.3.0` |
|
```yaml |
|
base_model: ./TinyLlama-1.1B-intermediate-step-1431k-3T |
|
|
|
model_type: LlamaForCausalLM |
|
tokenizer_type: LlamaTokenizer |
|
is_llama_derived_model: true |
|
|
|
load_in_8bit: false |
|
load_in_4bit: false |
|
strict: false |
|
|
|
datasets: |
|
- path: ./openhermes |
|
type: alpaca |
|
dataset_prepared_path: |
|
val_set_size: 0.05 |
|
output_dir: ./out |
|
|
|
sequence_len: 4096 |
|
sample_packing: false |
|
|
|
adapter: |
|
lora_model_dir: |
|
lora_r: |
|
lora_alpha: |
|
lora_dropout: |
|
lora_target_linear: |
|
lora_fan_in_fan_out: |
|
|
|
wandb_project: tinyllama-openhermes |
|
wandb_entity: tensoic |
|
wandb_watch: |
|
wandb_name: |
|
wandb_log_model: |
|
|
|
gradient_accumulation_steps: 2 |
|
micro_batch_size: 8 |
|
num_epochs: 1 |
|
optimizer: adamw_bnb_8bit |
|
lr_scheduler: cosine |
|
learning_rate: 0.0002 |
|
|
|
train_on_inputs: false |
|
group_by_length: false |
|
bf16: false |
|
fp16: true |
|
tf32: false |
|
|
|
gradient_checkpointing: true |
|
early_stopping_patience: |
|
resume_from_checkpoint: |
|
local_rank: |
|
logging_steps: 1 |
|
xformers_attention: true |
|
flash_attention: |
|
|
|
warmup_steps: 100 |
|
evals_per_epoch: 4 |
|
eval_table_size: |
|
saves_per_epoch: 1 |
|
debug: |
|
deepspeed: zero2.json |
|
weight_decay: 0.0 |
|
fsdp: |
|
fsdp_config: |
|
special_tokens: |
|
bos_token: "<s>" |
|
eos_token: "</s>" |
|
unk_token: "<unk>" |
|
|
|
``` |
|
|
|
</details><br> |
|
|
|
|
|
The loss for the 3T checkpoint explodes for some reason |
|
 |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 0.0002 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 8 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 128 |
|
- total_eval_batch_size: 64 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 100 |
|
- num_epochs: 1 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 3.0006 | 0.0 | 1 | 1.6838 | |
|
| 0.855 | 0.25 | 451 | 1.5228 | |
|
| 6.8636 | 0.5 | 902 | 7.4147 | |
|
| 6.9346 | 0.75 | 1353 | 7.4061 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.2 |
|
- Pytorch 2.0.1+cu117 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
