INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

advertisement

-0.75

jer

-0.70

comments

-0.68

 clicking

-0.67

 clicked

-0.66

 Gould

-0.66

 interfering

-0.65

Poe

-0.64

 CODE

-0.63

 misogyn

-0.63

POSITIVE LOGITS

llor

0.79

ath

0.77

iland

0.71

glers

0.70

yk

0.70

ysis

0.69

cycl

0.69

mar

0.69

 enthusi

0.68

 contrace

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.