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

Allah

-0.79

angan

-0.77

oths

-0.75

abama

-0.71

ulhu

-0.69

 Chaser

-0.69

 Nation

-0.69

NetMessage

-0.68

Balt

-0.66

 Guerrero

-0.66

POSITIVE LOGITS

PID

0.73

 Photograph

0.71

shape

0.67

ROM

0.67

sequ

0.66

 Instruments

0.66

 LIVE

0.64

RAM

0.63

binary

0.63

fit

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.