See axolotl config
axolotl version: 0.4.1
# See:
# - https://github.com/karpathy/nanoGPT/blob/master/config/train_gpt2.py#L1
# - https://github.com/OpenAccess-AI-Collective/axolotl/blob/main/examples/tiny-llama/pretrain.yml#L14
# - https://github.com/karpathy/nanoGPT/blob/master/train.py#L35
base_model: diwank/cryptgpt
hub_model_id: diwank/cryptgpt
model_type: GPT2LMHeadModel
tokenizer_type: AutoTokenizer
trust_remote_code: true # required for CryptGPTTokenizer
resize_token_embeddings_to_32x: true
output_dir: ./outputs/model-out
datasets:
- path: diwank/encrypted-openwebtext
type: completion
dataset_prepared_path: ./cryptgpt-prepared-dataset
val_set_size: 0.04
shuffle_merged_datasets: false
sequence_len: 1024
pad_to_sequence_len: true
sample_packing: false
pretrain_multipack_attn: false
train_on_inputs: true
gradient_accumulation_steps: 1
micro_batch_size: 64
optimizer: adamw_bnb_8bit
adam_beta1: 0.9
adam_beta2: 0.95
seed: 42
lr_scheduler: cosine
learning_rate: 6e-4
cosine_min_lr_ratio: 0.1 # min: 6e-5
weight_decay: 0.1
bf16: auto
tf32: true
flash_attention: true
torch_compile: true
gradient_checkpointing: false
deepspeed: deepspeed_configs/zero2.json
max_steps: 1200000
eval_steps: 12000
save_steps: 12000
auto_resume_from_checkpoints: true
logging_steps: 1
eval_max_new_tokens: 128
eval_causal_lm_metrics:
- sacrebleu
wandb_project: cryptgpt-0.1
wandb_name: cryptgpt-run-07
cryptgpt
This model is a fine-tuned version of diwank/cryptgpt on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.2717
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: 0.0006
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 512
- total_eval_batch_size: 512
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- training_steps: 40912
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
10.9453 | 0.0000 | 1 | 10.9383 |
3.0117 | 0.2933 | 12000 | 2.8623 |
2.5234 | 0.5866 | 24000 | 2.4040 |
2.3398 | 0.8799 | 36000 | 2.2717 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.1.2+cu118
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 32
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for diwank/cryptgpt
Unable to build the model tree, the base model loops to the model itself. Learn more.