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

stocks

-0.78

 seiz

-0.75

cape

-0.74

mitted

-0.69

letters

-0.65

doing

-0.65

inis

-0.64

inations

-0.64

anski

-0.64

DI

-0.64

POSITIVE LOGITS

unch

0.74

atable

0.66

ormons

0.65

lehem

0.63

 Murray

0.62

 Mehran

0.62

Utt

0.61

 Leah

0.61

ku

0.61

Him

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.