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

redo

-1.09

tein

-0.99

oggle

-0.89

poon

-0.78

lio

-0.77

eki

-0.77

rand

-0.76

arthed

-0.74

eal

-0.72

lander

-0.71

POSITIVE LOGITS

 approximate

0.62

ivan

0.61

 wors

0.60

 unim

0.60

rawdownloadcloneembedreportprint

0.60

Sus

0.60

 -------

0.59

bir

0.59

ript

0.58

 numbered

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.