INDEX

Explanations

No Explanations Found

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

ĪĴ

-0.81

seys

-0.80

chie

-0.79

rid

-0.70

 comprom

-0.70

rule

-0.70

igl

-0.70

ered

-0.69

chieve

-0.69

venge

-0.69

POSITIVE LOGITS

interstitial

0.74

çīĪ

0.65

 Amos

0.65

 Oslo

0.65

Via

0.62

Delta

0.62

baum

0.61

debian

0.60

pg

0.60

anka

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.