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

eret

-0.72

LC

-0.67

istor

-0.64

 cler

-0.64

NER

-0.63

quer

-0.62

 waitress

-0.62

 rapist

-0.61

iott

-0.60

ihad

-0.60

POSITIVE LOGITS

phabet

0.68

 symmetry

0.65

ategories

0.65

utsche

0.63

 Advantage

0.63

 istg

0.62

fun

0.62

 reminis

0.62

 Yahoo

0.62

abama

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.