INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Hour

-0.82

âķĲâķĲ

-0.77

âķĲ

-0.72

oufl

-0.70

aldehyde

-0.68

 Flore

-0.67

 Hopkins

-0.67

lv

-0.66

alion

-0.65

¶ħ

-0.64

POSITIVE LOGITS

istrate

0.87

 planner

0.81

istrates

0.74

ilater

0.71

ited

0.70

itor

0.67

itte

0.66

ophys

0.65

utive

0.64

 sucker

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.