Downloads · 30 days
14
34% of all-time downloads
llama_CN_pretrain
VikrantRamesh/llama_CN_pretrain
llama_CN_pretrain is a feature extraction model from VikrantRamesh. Use it when you need embeddings to search or compare text. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
41
Public
Parameters
6.8B
7.4 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors7.4 GB · 100%
Parameter types
How the weights are stored.
U85.6B · 82%
Base models
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
Llama_CN_pretrain
This model is a fine-tuned version of meta-llama/Llama-2-7b-chat-hf on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.5331
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 100
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.0225 | 0.03 | 10 | 3.1022 |
| 2.8635 | 0.05 | 20 | 2.7898 |
| 2.6242 | 0.08 | 30 | 2.7440 |
| 2.6995 | 0.1 | 40 | 2.6854 |
| 2.4791 | 0.13 | 50 | 2.6426 |
| 2.5932 | 0.15 | 60 | 2.5892 |
| 2.4885 | 0.18 | 70 | 2.5631 |
| 2.494 | 0.21 | 80 | 2.5508 |
| 2.4346 | 0.23 | 90 | 2.5361 |
| 2.4207 | 0.26 | 100 | 2.5331 |
Framework versions
- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Task
- Feature Extraction
- Library
- transformers
- Type
- llama
- Created
- Mar 17, 2024
- Updated
- Mar 18, 2024