INDEX

Explanations

dates or locations from news articles

oai_token-act-pair · gpt-3.5-turbo

references to "New" in various contexts and locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

stood

-0.76

ppe

-0.74

osate

-0.72

agn

-0.70

uddin

-0.68

mop

-0.68

abet

-0.65

ascript

-0.64

heit

-0.63

aminer

-0.63

POSITIVE LOGITS

 YORK

1.29

foundland

1.15

NEW

1.03

PORT

1.02

CAST

0.89

 Orleans

0.86

TERN

0.84

 York

0.83

 Zealand

0.82

ARK

0.82

Activations Density 0.005%

dates or locations from news articles

references to "New" in various contexts and locations

No Comments

No Known Activations