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

ãĥ¼ãĥĨãĤ£

-0.77

lihood

-0.73

sembly

-0.71

collar

-0.69

lder

-0.68

oun

-0.65

 rejoice

-0.65

ryu

-0.64

olicy

-0.64

ranch

-0.64

POSITIVE LOGITS

rary

0.69

ena

0.68

 Venezuel

0.67

egal

0.66

enezuel

0.65

Vul

0.64

myra

0.63

Uz

0.62

ifying

0.62

 OPEC

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.