INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

................................

-0.67

``(

-0.61

rawdownloadcloneembedreportprint

-0.61

 belts

-0.61

isSpecialOrderable

-0.60

 Agency

-0.60

 teeth

-0.59

 mandate

-0.58

gency

-0.58

++++++++++++++++

-0.58

POSITIVE LOGITS

lio

0.85

ocument

0.77

ipel

0.75

 Celt

0.69

essim

0.66

ructose

0.66

oulos

0.65

ividual

0.64

redd

0.63

vity

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.