INDEX

Explanations

mentions of the TV show "Survivor"

oai_token-act-pair · gpt-3.5-turbo

references to the TV show "Survivor" and its contestants

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

forward

-0.76

 pains

-0.75

 Stur

-0.74

 ware

-0.72

 brid

-0.70

ensing

-0.69

heter

-0.67

rity

-0.66

anical

-0.66

 resent

-0.66

POSITIVE LOGITS

 contestants

1.34

 contestant

1.33

 Solitaire

1.13

 Survivor

1.09

 Contest

1.02

 Apprentice

0.96

Surv

0.92

ipedia

0.87

achel

0.83

 Simulator

0.82

Activations Density 0.025%

mentions of the TV show "Survivor"

references to the TV show "Survivor" and its contestants

No Comments

No Known Activations