INDEX

Explanations

questions starting with "Why" or expressing confusion about a situation

oai_token-act-pair · gpt-3.5-turbo

rhetorical questions or expressions of disbelief

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

catentry

-0.93

xon

-0.69

 premie

-0.64

bye

-0.59

EStreamFrame

-0.58

Bye

-0.58

 firsthand

-0.57

illation

-0.56

 unmist

-0.55

 Goodbye

-0.55

POSITIVE LOGITS

!?"

0.84

?!"

0.83

 conceivable

0.83

 ever

0.82

 anyone

0.82

 EVER

0.81

 bother

0.81

 anybody

0.80

ANY

0.79

?!

0.77

Activations Density 0.305%

questions starting with "Why" or expressing confusion about a situation

rhetorical questions or expressions of disbelief

No Comments

No Known Activations