INDEX

Explanations

names related to the word "Swan" or variations of it

oai_token-act-pair · gpt-3.5-turbo

mentions of the names "Swan," "Swansea," and "Frozen."

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

uracy

-0.81

iating

-0.75

inez

-0.71

iveness

-0.71

Uz

-0.71

orate

-0.70

 dawn

-0.69

uala

-0.68

ICAN

-0.68

ARDS

-0.68

POSITIVE LOGITS

 Swan

1.34

 Rapids

0.79

Voy

0.76

 Harbour

0.76

 Swanson

0.72

daq

0.72

Fey

0.72

farm

0.72

Wem

0.72

Ĥª

0.71

Activations Density 0.007%

names related to the word "Swan" or variations of it

mentions of the names "Swan," "Swansea," and "Frozen."

No Comments

No Known Activations