Update README.md
Browse files
README.md
CHANGED
|
@@ -1,7 +1,12 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
|
| 5 |
metrics:
|
| 6 |
- accuracy
|
| 7 |
model-index:
|
|
@@ -9,9 +14,6 @@ model-index:
|
|
| 9 |
results: []
|
| 10 |
---
|
| 11 |
|
| 12 |
-
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
-
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
-
|
| 15 |
# openchatgpt-neo-r1
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [EleutherAI/gpt-neo-125M](https://huggingface.co/EleutherAI/gpt-neo-125M) on the openchatgpt safe-r1 dataset.
|
|
@@ -21,18 +23,26 @@ It achieves the following results on the evaluation set:
|
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
| 24 |
-
|
|
|
|
|
|
|
| 25 |
|
| 26 |
## Intended uses & limitations
|
| 27 |
|
| 28 |
-
|
|
|
|
|
|
|
| 29 |
|
| 30 |
## Training and evaluation data
|
| 31 |
|
| 32 |
-
|
|
|
|
|
|
|
| 33 |
|
| 34 |
## Training procedure
|
| 35 |
|
|
|
|
|
|
|
| 36 |
### Training hyperparameters
|
| 37 |
|
| 38 |
The following hyperparameters were used during training:
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
| 7 |
+
- text generation
|
| 8 |
+
- pytorch
|
| 9 |
+
- casual-lm
|
| 10 |
metrics:
|
| 11 |
- accuracy
|
| 12 |
model-index:
|
|
|
|
| 14 |
results: []
|
| 15 |
---
|
| 16 |
|
|
|
|
|
|
|
|
|
|
| 17 |
# openchatgpt-neo-r1
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [EleutherAI/gpt-neo-125M](https://huggingface.co/EleutherAI/gpt-neo-125M) on the openchatgpt safe-r1 dataset.
|
|
|
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
| 26 |
+
Finetune based on the inner workings of ChatGPT. I won't elaborate on that. You must have a faint idea of how prompt is made for it to spit anything that's not garbled mess.
|
| 27 |
+
|
| 28 |
+
This is effectively a schizophrenic idea that met the light of day. Practically a collab of 3 students in a virtual shed.
|
| 29 |
|
| 30 |
## Intended uses & limitations
|
| 31 |
|
| 32 |
+
Intended uses & limitations fall in line with OpenAI's. Dataset used consists of safe texts (i.e. not highly sexual/erotica type stuff). NSFW version of the dataset is not planned to exist at the moment.
|
| 33 |
+
|
| 34 |
+
Keep in mind that this is a 125m version of GPT-Neo. My 1050Ti Mobile couldn't even handle that without gradient thingmabobs. If anyone knows how to effectively finetune larger models on free colabs - feel free to let me know. Pile tokenizer also has one downside compared to native GPT-2/3 - `Assistant`.
|
| 35 |
|
| 36 |
## Training and evaluation data
|
| 37 |
|
| 38 |
+
Data was split in ratio of 95%/5%. Preproccess included removing mentions of OpenAI wherever it was not deemed appropriete (GPT-2 has one of the appropriete mentions). Whole dataset consists of just shy off 3k input-output pairs. One input has multiple outputs (read as: one message has multiple variants of an answer). <<<1% (3 total) are curated lines (i.e. a huge mistake was spotted that needed corrections).
|
| 39 |
+
|
| 40 |
+
Heavy bias on IT.
|
| 41 |
|
| 42 |
## Training procedure
|
| 43 |
|
| 44 |
+
Input and output were straight up concatenated due to the nature of how ChatGPT works. Padding chosen was the same as the separator token, if that's not effective - please let me know as I am new to this stuff.
|
| 45 |
+
|
| 46 |
### Training hyperparameters
|
| 47 |
|
| 48 |
The following hyperparameters were used during training:
|