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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Zone

-0.80

 Gund

-0.76

artisan

-0.75

 Corps

-0.66

 Gleaming

-0.65

eland

-0.65

XV

-0.65

 Aerial

-0.65

 Earth

-0.65

Caf

-0.64

POSITIVE LOGITS

 sche

0.84

UGC

0.69

osph

0.69

soever

0.66

kell

0.65

 fuelled

0.62

conn

0.60

 harbour

0.58

 comprising

0.58

 motivated

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.