INDEX

Explanations

instances where doubt is expressed towards a certain idea or statement

oai_token-act-pair · gpt-3.5-turbo

expressions of skepticism or uncertainty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ramid

-0.82

emetery

-0.78

alez

-0.72

zie

-0.71

ibrary

-0.71

ossier

-0.69

ummer

-0.68

insula

-0.67

apeake

-0.67

ocument

-0.66

POSITIVE LOGITS

lessly

1.33

worthiness

0.96

fully

0.92

ingly

0.86

fulness

0.82

 doubt

0.77

imaru

0.77

lessness

0.77

 doubts

0.71

 hesitation

0.69

Activations Density 0.018%

instances where doubt is expressed towards a certain idea or statement

expressions of skepticism or uncertainty

No Comments

No Known Activations