metadata
library_name: transformers
base_model: trl-internal-testing/tiny-random-LlamaForCausalLM
tags:
- axolotl
- generated_from_trainer
datasets:
- argilla/databricks-dolly-15k-curated-en
model-index:
- name: tiny-random-LlamaForCausalLM
results: []
See axolotl config
axolotl version: 0.6.0
base_model: trl-internal-testing/tiny-random-LlamaForCausalLM
batch_size: 64
bf16: true
chat_template: tokenizer_default_fallback_alpaca
datasets:
- format: custom
path: argilla/databricks-dolly-15k-curated-en
type:
field_input: original-instruction
field_instruction: original-instruction
field_output: original-response
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
device_map: auto
eval_sample_packing: false
eval_steps: 40
flash_attention: true
gradient_checkpointing: true
group_by_length: true
hub_model_id: SystemAdmin123/tiny-random-LlamaForCausalLM
hub_strategy: checkpoint
learning_rate: 0.0002
logging_steps: 10
lr_scheduler: cosine
max_steps: 5000
micro_batch_size: 32
model_type: AutoModelForCausalLM
num_epochs: 100
optimizer: adamw_bnb_8bit
output_dir: /root/.sn56/axolotl/tmp/tiny-random-LlamaForCausalLM
pad_to_sequence_len: true
resize_token_embeddings_to_32x: false
sample_packing: true
save_steps: 20
save_total_limit: 2
sequence_len: 2048
tokenizer_type: LlamaTokenizerFast
torch_dtype: bf16
trust_remote_code: true
val_set_size: 0.1
wandb_entity: ''
wandb_mode: online
wandb_name: trl-internal-testing/tiny-random-LlamaForCausalLM-argilla/databricks-dolly-15k-curated-en
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: default
warmup_ratio: 0.05
tiny-random-LlamaForCausalLM
This model is a fine-tuned version of trl-internal-testing/tiny-random-LlamaForCausalLM on the argilla/databricks-dolly-15k-curated-en dataset. It achieves the following results on the evaluation set:
- Loss: 9.1944
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: 32
- eval_batch_size: 32
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 64
- total_eval_batch_size: 64
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 30
- training_steps: 600
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0.0769 | 1 | 10.3764 |
| 10.3522 | 3.0769 | 40 | 10.3366 |
| 10.1177 | 6.1538 | 80 | 10.0885 |
| 9.8887 | 9.2308 | 120 | 9.8677 |
| 9.688 | 12.3077 | 160 | 9.6724 |
| 9.5151 | 15.3846 | 200 | 9.5050 |
| 9.3725 | 18.4615 | 240 | 9.3687 |
| 9.2678 | 21.5385 | 280 | 9.2734 |
| 9.2101 | 24.6154 | 320 | 9.2205 |
| 9.186 | 27.6923 | 360 | 9.2018 |
| 9.18 | 30.7692 | 400 | 9.1964 |
| 9.1787 | 33.8462 | 440 | 9.1945 |
| 9.1768 | 36.9231 | 480 | 9.1941 |
| 9.1775 | 40.0 | 520 | 9.1938 |
| 9.1784 | 43.0769 | 560 | 9.1949 |
| 9.1762 | 46.1538 | 600 | 9.1944 |
Framework versions
- Transformers 4.48.1
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0