INDEX

Explanations

descriptions related to locations or attractions

oai_token-act-pair · gpt-3.5-turbo

references to visual experiences or sights

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mand

-0.80

 mandate

-0.77

reporting

-0.70

CLAIM

-0.70

Correction

-0.67

pec

-0.66

imester

-0.64

jet

-0.64

aniel

-0.63

tip

-0.63

POSITIVE LOGITS

 goodies

1.06

 treasures

0.98

 firsthand

0.98

 splend

0.94

 scenery

0.93

 classics

0.91

 attractions

0.91

abilia

0.88

 exciting

0.87

 amazing

0.86

Activations Density 0.917%

descriptions related to locations or attractions

references to visual experiences or sights

No Comments

No Known Activations