INDEX

Explanations

words related to locations or places

oai_token-act-pair · gpt-3.5-turbo

the occurrence of the substring "st" within various contexts in the text

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

hound

-0.83

perty

-0.79

merce

-0.73

gdala

-0.73

 largeDownload

-0.69

ggies

-0.67

prus

-0.65

aceous

-0.64

ITED

-0.64

iggurat

-0.63

POSITIVE LOGITS

oppers

0.98

alker

0.90

retch

0.89

amped

0.87

utter

0.85

eal

0.83

opped

0.83

Í

0.83

ealing

0.82

rict

0.82

Activations Density 0.016%

words related to locations or places

the occurrence of the substring "st" within various contexts in the text

No Comments

No Known Activations