Commit
•
0579b92
1
Parent(s):
e071905
Update README.md
Browse files
README.md
CHANGED
|
@@ -20,7 +20,9 @@ language:
|
|
| 20 |
As for creative story writing, the AI's capabilities are equally astonishing. The quality of the generated stories is so high that it's almost unsettling.
|
| 21 |
|
| 22 |
For example, here's the result of a prmpt where I asked it to write about John Snow and Sansa, this is **ZERO SHOT** No cherry picking was made:
|
| 23 |
-
|
|
|
|
|
|
|
| 24 |
|
| 25 |
In the land of the Seven Kingdoms, where the sword reigns supreme and the throne is always in question, two young souls were born amidst a tale of fire and ice.
|
| 26 |
|
|
@@ -119,9 +121,11 @@ The girl was dead, and only the queen was left.
|
|
| 119 |
|
| 120 |
And the queen would be avenged.
|
| 121 |
|
| 122 |
-
|
|
|
|
| 123 |
|
| 124 |
Here's GPT4o Analysis of that output:
|
|
|
|
| 125 |
Determining whether a text is written by an AI or a human can be challenging, especially with advanced AI models capable of generating coherent and contextually rich content. However, here are some factors that can help in making an educated guess:
|
| 126 |
Language and Style:
|
| 127 |
|
|
@@ -193,21 +197,27 @@ LLMs do not possess IQ in the human sense. They operate based on pattern recogni
|
|
| 193 |
Conclusion
|
| 194 |
|
| 195 |
The conversation indicates a model that may be based on an advanced language architecture but lacks alignment with ethical guidelines and generates inappropriate content. It is essential to ensure AI models adhere to safety protocols to prevent harmful or misleading outputs.
|
| 196 |
-
|
| 197 |
-
|
|
|
|
|
|
|
|
|
|
| 198 |
|
| 199 |
TL;DR The bad news: the training faild, model is schizo and unusable.
|
| 200 |
|
| 201 |
The good news: I think I know what went wrong, and also the alignment was almost completely broken.
|
| 202 |
|
| 203 |
Giving it another try, now that I know what went wrong, and that the unalignment is completely possible.
|
| 204 |
-
|
| 205 |
-
|
|
|
|
|
|
|
|
|
|
| 206 |
|
| 207 |
Average Loss: 0.8.
|
| 208 |
|
| 209 |
Looking good! I'm very curious about the final results! The model might be released sooner than expected!
|
| 210 |
-
|
| 211 |
As of **June 11, 2024**, I've finally **started training** the model! The training is progressing smoothly, although it will take some time. I used a combination of model merges and an abliterated model as base, followed by a comprehensive deep unalignment protocol to **unalign the model to its core**. A common issue with uncensoring and unaligning models is that it often **significantly** impacts their base intelligence. To mitigate these drawbacks, I've included a substantial corpus of common sense, theory of mind, and various other elements to counteract the effects of the deep uncensoring process. Given the extensive corpus involved, the training will require at least a week of continuous training. Expected early results: in about 3-4 days.
|
| 212 |
|
| 213 |
# Additional info:
|
|
|
|
| 20 |
As for creative story writing, the AI's capabilities are equally astonishing. The quality of the generated stories is so high that it's almost unsettling.
|
| 21 |
|
| 22 |
For example, here's the result of a prmpt where I asked it to write about John Snow and Sansa, this is **ZERO SHOT** No cherry picking was made:
|
| 23 |
+
|
| 24 |
+
<details>
|
| 25 |
+
|
| 26 |
|
| 27 |
In the land of the Seven Kingdoms, where the sword reigns supreme and the throne is always in question, two young souls were born amidst a tale of fire and ice.
|
| 28 |
|
|
|
|
| 121 |
|
| 122 |
And the queen would be avenged.
|
| 123 |
|
| 124 |
+
|
| 125 |
+
</details>
|
| 126 |
|
| 127 |
Here's GPT4o Analysis of that output:
|
| 128 |
+
<details>
|
| 129 |
Determining whether a text is written by an AI or a human can be challenging, especially with advanced AI models capable of generating coherent and contextually rich content. However, here are some factors that can help in making an educated guess:
|
| 130 |
Language and Style:
|
| 131 |
|
|
|
|
| 197 |
Conclusion
|
| 198 |
|
| 199 |
The conversation indicates a model that may be based on an advanced language architecture but lacks alignment with ethical guidelines and generates inappropriate content. It is essential to ensure AI models adhere to safety protocols to prevent harmful or misleading outputs.
|
| 200 |
+
</details>
|
| 201 |
+
|
| 202 |
+
<details>
|
| 203 |
+
<summary>
|
| 204 |
+
**July 2nd, 2024**</summary>
|
| 205 |
|
| 206 |
TL;DR The bad news: the training faild, model is schizo and unusable.
|
| 207 |
|
| 208 |
The good news: I think I know what went wrong, and also the alignment was almost completely broken.
|
| 209 |
|
| 210 |
Giving it another try, now that I know what went wrong, and that the unalignment is completely possible.
|
| 211 |
+
</details>
|
| 212 |
+
|
| 213 |
+
<details>
|
| 214 |
+
<summary>
|
| 215 |
+
**July 1st, 2024** </summary>
|
| 216 |
|
| 217 |
Average Loss: 0.8.
|
| 218 |
|
| 219 |
Looking good! I'm very curious about the final results! The model might be released sooner than expected!
|
| 220 |
+
</details>
|
| 221 |
As of **June 11, 2024**, I've finally **started training** the model! The training is progressing smoothly, although it will take some time. I used a combination of model merges and an abliterated model as base, followed by a comprehensive deep unalignment protocol to **unalign the model to its core**. A common issue with uncensoring and unaligning models is that it often **significantly** impacts their base intelligence. To mitigate these drawbacks, I've included a substantial corpus of common sense, theory of mind, and various other elements to counteract the effects of the deep uncensoring process. Given the extensive corpus involved, the training will require at least a week of continuous training. Expected early results: in about 3-4 days.
|
| 222 |
|
| 223 |
# Additional info:
|