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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 remotely

-0.69

 Baton

-0.64

nob

-0.64

erence

-0.62

aires

-0.62

 Notting

-0.62

 Persona

-0.62

mate

-0.62

Ake

-0.61

 pist

-0.59

POSITIVE LOGITS

WAR

0.77

umn

0.75

iatrics

0.72

oufl

0.70

MM

0.69

stros

0.68

misc

0.68

RR

0.68

thur

0.65

————

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.