INDEX

Explanations

references to the name "Thor" specifically

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Thor."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

URES

-0.76

 sterling

-0.72

yles

-0.66

EAR

-0.66

holders

-0.65

assetsadobe

-0.64

pring

-0.63

aminer

-0.63

 coerc

-0.62

eva

-0.61

POSITIVE LOGITS

 Ragnarok

1.13

sten

1.12

nton

1.11

bj

0.93

azine

0.90

ax

0.86

mand

0.85

chest

0.85

aldi

0.81

ium

0.81

Activations Density 0.047%

references to the name "Thor" specifically

mentions of the name "Thor."

No Comments

No Known Activations