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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

athering

-0.71

ighters

-0.68

ints

-0.67

ancock

-0.62

ĻĤ

-0.60

 magistrate

-0.59

ooting

-0.59

pron

-0.58

ĸļ

-0.58

essler

-0.57

POSITIVE LOGITS

Msg

0.65

awa

0.65

 abroad

0.62

Pos

0.62

goal

0.62

ãĤ¦

0.61

race

0.60

 Guant

0.59

 exempted

0.59

ombat

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.