INDEX

Explanations

No Explanations Found

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

clude

-0.75

nell

-0.73

luster

-0.72

ppo

-0.72

lege

-0.71

clusively

-0.69

loading

-0.68

cluding

-0.66

quire

-0.65

 Dawkins

-0.64

POSITIVE LOGITS

 inspectors

0.82

hran

0.81

acas

0.78

 Hague

0.73

acan

0.73

oÄŁ

0.71

amaru

0.69

 negotiators

0.67

taboola

0.67

 Democr

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.