INDEX

Explanations

mentions of hindrance or obstruction

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "hind."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Rated

-0.77

atis

-0.74

itÃ©

-0.74

terday

-0.74

atically

-0.74

mson

-0.71

å£«

-0.67

ħĭ

-0.65

ailability

-0.65

ably

-0.63

POSITIVE LOGITS

ering

1.60

erer

1.47

erers

1.47

ered

1.41

rance

1.20

ern

1.08

ers

1.03

ezvous

0.96

quarters

0.95

erest

0.90

Activations Density 0.057%

mentions of hindrance or obstruction

occurrences of the word "hind."

No Comments

No Known Activations