INDEX

Explanations

mentions or descriptions of hats

oai_token-act-pair · gpt-3.5-turbo

references to hats

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 UNIVERS

-0.70

 Rhod

-0.70

 PowerPoint

-0.64

MSM

-0.62

 glutamate

-0.62

«ĺ

-0.61

 Supplementary

-0.61

 TEST

-0.61

 Snapdragon

-0.60

Interstitial

-0.60

POSITIVE LOGITS

chet

1.30

chery

1.23

hat

1.22

 hats

1.11

cher

0.98

glers

0.92

ocular

0.92

Hat

0.89

gey

0.89

Hat

0.87

Activations Density 0.010%

mentions or descriptions of hats

references to hats

No Comments

No Known Activations