Update README.md
Browse files
README.md
CHANGED
|
@@ -31,7 +31,7 @@ BRAINSTORM process was developed by David_AU.
|
|
| 31 |
|
| 32 |
<B>What is Brainstorm?</b>
|
| 33 |
|
| 34 |
-
The reasoning center of an LLM is taken apart, reassembled, expanded and multiplied by 5x and 10x respectively. Then these are individually calibrated.
|
| 35 |
This process occurs at the root level - source files level. It does not use Imatrix, X-quants, Neo Class datasets and it not dependant
|
| 36 |
on LLAMA.cpp per say.
|
| 37 |
|
|
@@ -59,6 +59,7 @@ More testing is underway and there are additional sizes beyond "5x" and "10x".
|
|
| 59 |
- Sense of "there" is enhanced.
|
| 60 |
- Details are more vivid.
|
| 61 |
- Prose generation length can be extreme.
|
|
|
|
| 62 |
- These quants are very sensitive to "temp" and "rep-pen" changes - adjust by .01 or less each.
|
| 63 |
- Suggest a temp of .2 to start, rep pen of 1.1
|
| 64 |
- Although no issues have been detected during testing, they could still occur.
|
|
|
|
| 31 |
|
| 32 |
<B>What is Brainstorm?</b>
|
| 33 |
|
| 34 |
+
The reasoning center of an LLM is taken apart, reassembled, expanded and multiplied by 5x and 10x respectively. Then these centers are individually calibrated.
|
| 35 |
This process occurs at the root level - source files level. It does not use Imatrix, X-quants, Neo Class datasets and it not dependant
|
| 36 |
on LLAMA.cpp per say.
|
| 37 |
|
|
|
|
| 59 |
- Sense of "there" is enhanced.
|
| 60 |
- Details are more vivid.
|
| 61 |
- Prose generation length can be extreme.
|
| 62 |
+
- The MORE instructions and/or details you provide the more strongly the model will respond.
|
| 63 |
- These quants are very sensitive to "temp" and "rep-pen" changes - adjust by .01 or less each.
|
| 64 |
- Suggest a temp of .2 to start, rep pen of 1.1
|
| 65 |
- Although no issues have been detected during testing, they could still occur.
|