INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

agements

-0.68

Emer

-0.68

\\\\\\\\\\\\\\\\

-0.66

natureconservancy

-0.65

inges

-0.65

acles

-0.65

Sus

-0.64

ogle

-0.62

————————————————

-0.62

gement

-0.61

POSITIVE LOGITS

inois

0.71

calling

0.69

ilon

0.66

 criticizing

0.66

zai

0.65

CPC

0.64

olin

0.62

trak

0.62

senal

0.62

fascist

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.