NX-AI/xLSTM-7b
Updated
•
9.12k
•
62
Max Current
eldogbbhed
AI & ML interests
None yet
Recent Activity
liked
a model
4 days ago
NX-AI/xLSTM-7b
liked
a model
4 days ago
moxin-org/moxin-chat-7b
liked
a model
6 days ago
bigcode/starcoder2-7b
Organizations
None yet
eldogbbhed's activity
liked
2
models
4 days ago
liked
a
model
6 days ago
liked
a
model
14 days ago
liked
3
models
2 months ago
liked
a
Space
3 months ago
reacted to
Smooke's
post with 🧠
3 months ago
Post
601
Chomsky predicting LLMs in 1956, curated by Ryan Rhodes (Rutgers)
liked
6
models
3 months ago
FPHam/L3-8B-Everything-COT
Text Generation
•
Updated
•
141
•
15
kyutai/moshiko-candle-bf16
Updated
•
410
•
11
kyutai/moshika-pytorch-bf16
Updated
•
267
•
45
Exthalpy/state-0
Text Generation
•
Updated
•
4
•
12
ICTNLP/Llama-3.1-8B-Omni
Updated
•
5.34k
•
384
arcee-ai/Llama-3.1-SuperNova-Lite
Text Generation
•
Updated
•
6.72k
•
182
liked
3
models
4 months ago
reacted to
mlabonne's
post with 👍
4 months ago
Post
17323
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
liked
a
model
4 months ago
