metadata

license: other
base_model: meta-llama/Meta-Llama-3-70B
tags:
  - llama-factory
  - full
  - generated_from_trainer
model-index:
  - name: C021_random_sample_Meta-Llama-3-70B_pretrain_20240512_020339
    results: []

C021_random_sample_Meta-Llama-3-70B_pretrain_20240512_020339

This model is a fine-tuned version of /mnt/fl/models/llama3/Meta-Llama-3-70B on the C021_random_sample_data dataset.

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: 3e-06
train_batch_size: 1
eval_batch_size: 2
seed: 42
distributed_type: multi-GPU
num_devices: 32
gradient_accumulation_steps: 4
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: polynomial
lr_scheduler_warmup_ratio: 0.075
num_epochs: 0.5
mixed_precision_training: Native AMP

Training results

Framework versions

Transformers 4.40.2
Pytorch 2.3.0
Datasets 2.19.1
Tokenizers 0.19.1