rubenamtz / law2entity

Turn text from the law in Spanish to structured data to feed support building a knowledge graph

  • Public
  • 2 runs
Run with an API
Playground API Examples README Versions

Run time and cost

This model runs on Nvidia A40 (Large) GPU hardware. We don't yet have enough runs of this model to provide performance information.

Readme

license: llama3 library_name: peft tags: - axolotl - generated_from_trainer base_model: meta-llama/Meta-Llama-3-8B model-index: - name: llama-3-8b-lora-law2entity results: [] datasets: - rubenamtz0/law_entity_recognition

Built with Axolotl

<details><summary>See axolotl config</summary> axolotl version: `0.4.1` 
base_model: meta-llama/Meta-Llama-3-8B
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: true
load_in_4bit: false
strict: false

datasets:
  - path: rubenamtz0/law_entity_recognition
    type: alpaca
dataset_prepared_path:
val_set_size: 0.1
output_dir: ./outputs/lora-law
hub_model_id: rubenamtz0/llama-3-8b-lora-law2entity

sequence_len: 4096
sample_packing: true
pad_to_sequence_len: true

adapter: lora
lora_model_dir:
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:

wandb_project: entity-relationship-claim-ft
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 4
optimizer: adamw_bnb_8bit
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
s2_attention:

warmup_steps: 10
evals_per_epoch: 4
eval_table_size:
eval_max_new_tokens: 128
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
   pad_token: <|end_of_text|>
</details>


llama-3-8b-lora-law2entity

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B on the rubenamtz0/law_entity_recognition dataset. It achieves the following results on the evaluation set: - Loss: 0.1490

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 - distributed_type: multi-GPU - num_devices: 3 - gradient_accumulation_steps: 4 - total_train_batch_size: 24 - total_eval_batch_size: 6 - 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
0.2735 0.05 1 0.2923
0.2852 0.25 5 0.2742
0.2007 0.5 10 0.2015
0.1742 0.75 15 0.1807
0.1854 1.0 20 0.1688
0.159 1.1125 25 0.1630
0.1444 1.3625 30 0.1592
0.1479 1.6125 35 0.1565
0.1505 1.8625 40 0.1538
0.1369 2.1125 45 0.1518
0.1348 2.2125 50 0.1512
0.1287 2.4625 55 0.1510
0.1359 2.7125 60 0.1498
0.1367 2.9625 65 0.1491
0.1218 3.075 70 0.1491
0.1285 3.325 75 0.1493
0.1307 3.575 80 0.1490

Framework versions

  • PEFT 0.11.1
  • Transformers 4.41.1
  • Pytorch 2.1.2+cu118
  • Datasets 2.19.1
  • Tokenizers 0.19.1