INDEX

Explanations

years mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

numerical data or specific dates

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

plom

-0.84

afort

-0.76

kell

-0.73

ablo

-0.71

Ã±o

-0.71

hei

-0.71

pora

-0.70

uana

-0.69

iful

-0.67

estic

-0.67

POSITIVE LOGITS

âĸĪâĸĪ

0.94

committee

0.84

th

0.77

SW

0.75

ISH

0.73

 minutes

0.72

0.71

mg

0.71

oz

0.71

Activations Density 0.042%

years mentioned in the text

numerical data or specific dates

No Comments

No Known Activations