INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

è¡

-0.69

Ð²

-0.67

bley

-0.64

RIC

-0.62

âĿ

-0.60

 spiral

-0.57

 Westminster

-0.56

È

-0.56

 Sounders

-0.56

ius

-0.56

POSITIVE LOGITS

 concess

0.73

itt

0.71

 Fisheries

0.70

object

0.68

 corros

0.68

conom

0.68

ailable

0.67

usat

0.66

options

0.66

ask

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.