INDEX

Explanations

negative emotional words, related to disappointment, disbelief, and critique

oai_token-act-pair · gpt-3.5-turbo

negative emotional responses or concepts related to disappointment and disillusionment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

sky

-0.72

ãĤ´ãĥ³

-0.70

minster

-0.69

ãģ®å®

-0.68

 Tomorrow

-0.65

 amen

-0.61

frey

-0.60

 doubling

-0.59

enium

-0.58

 Industry

-0.57

POSITIVE LOGITS

owship

0.93

essional

0.90

untled

0.87

ption

0.82

ibles

0.80

atives

0.80

isitions

0.76

ible

0.75

agement

0.74

ointed

0.74

Activations Density 0.065%

negative emotional words, related to disappointment, disbelief, and critique

negative emotional responses or concepts related to disappointment and disillusionment

No Comments

No Known Activations