story writing favourites
Collection
Models I personally liked for generating stories in the past. Not a recommendation, many of these are outdated.
•
17 items
•
Updated
•
29
Primer Primed
PrimeD
AI & ML interests
None yet
Recent Activity
upvoted
a
collection
22 days ago
story writing favourites
liked
a model
about 1 month ago
TheDrummer/Tunguska-39B-v1-GGUF
Organizations
None yet
PrimeD's activity
upvoted
a
collection
22 days ago
can we have GGUF for "creative-writer-v0.2-bravo-35b" ?
5
#1 opened 24 days ago
by
PrimeD
liked
a
model
about 1 month ago
IQ3_M Quant?
3
#2 opened about 2 months ago
by
PrimeD
Invalid Split File
6
#1 opened about 2 months ago
by
PrimeD
liked
a
model
about 2 months ago
liked
a
model
2 months ago
liked
a
model
3 months ago
[Feedback]
5
#1 opened 4 months ago
by
Darkknight535
liked
2
models
4 months ago
Would there be a version of model with bigger context size?
2
#1 opened 4 months ago
by
PrimeD
liked
a
model
4 months ago
Is this the updated version or the bugged version?
7
#2 opened 4 months ago
by
PrimeD
Thanks for the Update
1
#1 opened 4 months ago
by
PrimeD
liked
2
models
4 months ago
Do we have to unzip/ concatenate the gguf files before using them?
5
#1 opened 5 months ago
by
PrimeD
liked
a
model
5 months ago
reacted to
mlabonne's
post with 👍
6 months ago
Post
17521
Large models are surprisingly bad storytellers.
I asked 8 LLMs to "Tell me a bedtime story about bears and waffles."
Claude 3.5 Sonnet and GPT-4o gave me the worst stories: no conflict, no moral, zero creativity.
In contrast, smaller models were quite creative and wrote stories involving talking waffle trees and bears ostracized for their love of waffles.
Here you can see a comparison between Claude 3.5 Sonnet and NeuralDaredevil-8B-abliterated. They both start with a family of bears but quickly diverge in terms of personality, conflict, etc.
I mapped it to the hero's journey to have some kind of framework. Prompt engineering can definitely help here, but it's still disappointing that the larger models don't create better stories right off the bat.
Do you know why smaller models outperform the frontier models here?
I asked 8 LLMs to "Tell me a bedtime story about bears and waffles."
Claude 3.5 Sonnet and GPT-4o gave me the worst stories: no conflict, no moral, zero creativity.
In contrast, smaller models were quite creative and wrote stories involving talking waffle trees and bears ostracized for their love of waffles.
Here you can see a comparison between Claude 3.5 Sonnet and NeuralDaredevil-8B-abliterated. They both start with a family of bears but quickly diverge in terms of personality, conflict, etc.
I mapped it to the hero's journey to have some kind of framework. Prompt engineering can definitely help here, but it's still disappointing that the larger models don't create better stories right off the bat.
Do you know why smaller models outperform the frontier models here?
- 44 replies