INDEX

Explanations

phrases indicating causality

oai_token-act-pair · gpt-3.5-turbo

the word "so" used as a connector or transition in sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

mast

-0.62

 Mens

-0.61

ammy

-0.59

inch

-0.59

å¥³

-0.58

Wer

-0.56

 Halls

-0.56

kb

-0.55

 silhouette

-0.53

 Souls

-0.53

POSITIVE LOGITS

bered

1.22

oner

1.14

othe

1.14

apy

1.12

othes

0.99

aps

0.92

oooo

0.88

arin

0.87

ooo

0.85

oths

0.82

Activations Density 0.067%

phrases indicating causality

the word "so" used as a connector or transition in sentences

No Comments

No Known Activations