INDEX

Explanations

mentions of specific geographic locations, particularly national parks

oai_token-act-pair · gpt-3.5-turbo

references to bears and specific national parks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

inen

-0.89

oa

-0.86

 advoc

-0.83

ischer

-0.82

rums

-0.81

printf

-0.77

usc

-0.76

ohn

-0.76

estinal

-0.75

atta

-0.75

POSITIVE LOGITS

 Yosemite

0.92

 Grizz

0.85

 grizz

0.84

 Delta

0.83

 Yellowstone

0.78

 Grizzlies

0.76

 footh

0.73

 jumper

0.72

 Doodle

0.69

 delta

0.68

Activations Density 0.016%

mentions of specific geographic locations, particularly national parks

references to bears and specific national parks

No Comments

No Known Activations