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

 abroad

-0.77

cream

-0.68

 Trophy

-0.65

 themed

-0.63

flex

-0.61

 Ivanka

-0.60

 TRUMP

-0.60

 Gould

-0.59

 Lauder

-0.59

 Sovereign

-0.59

POSITIVE LOGITS

itsch

0.74

arten

0.72

iston

0.71

IG

0.69

aden

0.68

urity

0.66

 innocence

0.65

amins

0.65

atur

0.65

ari

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.