Model save
Browse files- README.md +76 -0
- adapter_model.bin +1 -1
README.md
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Llama-2-13b-hf
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: Llama-2-13b-hf-eli5-wiki-1024_qlora
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# Llama-2-13b-hf-eli5-wiki-1024_qlora
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [meta-llama/Llama-2-13b-hf](https://huggingface.co/meta-llama/Llama-2-13b-hf) on the None dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 1.3502
|
| 18 |
+
|
| 19 |
+
## Model description
|
| 20 |
+
|
| 21 |
+
More information needed
|
| 22 |
+
|
| 23 |
+
## Intended uses & limitations
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Training and evaluation data
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training procedure
|
| 32 |
+
|
| 33 |
+
### Training hyperparameters
|
| 34 |
+
|
| 35 |
+
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 0.0002
|
| 37 |
+
- train_batch_size: 16
|
| 38 |
+
- eval_batch_size: 16
|
| 39 |
+
- seed: 42
|
| 40 |
+
- gradient_accumulation_steps: 8
|
| 41 |
+
- total_train_batch_size: 128
|
| 42 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
+
- lr_scheduler_type: linear
|
| 44 |
+
- lr_scheduler_warmup_ratio: 0.03
|
| 45 |
+
- num_epochs: 3
|
| 46 |
+
|
| 47 |
+
### Training results
|
| 48 |
+
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 51 |
+
| 1.5071 | 0.15 | 125 | 1.3593 |
|
| 52 |
+
| 1.5606 | 0.3 | 250 | 1.3514 |
|
| 53 |
+
| 1.452 | 0.45 | 375 | 1.3441 |
|
| 54 |
+
| 1.5435 | 0.6 | 500 | 1.3423 |
|
| 55 |
+
| 1.4355 | 0.75 | 625 | 1.3373 |
|
| 56 |
+
| 1.5407 | 0.9 | 750 | 1.3371 |
|
| 57 |
+
| 1.1397 | 1.05 | 875 | 1.3361 |
|
| 58 |
+
| 1.4666 | 1.2 | 1000 | 1.3383 |
|
| 59 |
+
| 1.1651 | 1.35 | 1125 | 1.3374 |
|
| 60 |
+
| 1.4819 | 1.5 | 1250 | 1.3372 |
|
| 61 |
+
| 1.1667 | 1.65 | 1375 | 1.3361 |
|
| 62 |
+
| 1.4585 | 1.8 | 1500 | 1.3348 |
|
| 63 |
+
| 1.138 | 1.95 | 1625 | 1.3339 |
|
| 64 |
+
| 1.069 | 2.1 | 1750 | 1.3474 |
|
| 65 |
+
| 1.7055 | 2.25 | 1875 | 1.3596 |
|
| 66 |
+
| 1.0747 | 2.4 | 2000 | 1.3498 |
|
| 67 |
+
| 1.7118 | 2.55 | 2125 | 1.3580 |
|
| 68 |
+
| 1.0339 | 2.7 | 2250 | 1.3502 |
|
| 69 |
+
|
| 70 |
+
|
| 71 |
+
### Framework versions
|
| 72 |
+
|
| 73 |
+
- Transformers 4.34.0.dev0
|
| 74 |
+
- Pytorch 2.0.1+cu118
|
| 75 |
+
- Datasets 2.14.5
|
| 76 |
+
- Tokenizers 0.13.3
|
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1001592141
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f8331afb1b07b933475c6e7906f9be2e400240c5c5b807a9c968fd9187890b8
|
| 3 |
size 1001592141
|