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

iny

-0.73

 Lump

-0.70

heit

-0.69

thus

-0.69

 ç¥ŀ

-0.66

Kut

-0.65

 Bash

-0.64

Ruk

-0.64

ske

-0.64

 Contrast

-0.63

POSITIVE LOGITS

ectar

0.70

iband

0.69

ISTER

0.67

everal

0.65

Student

0.64

Ey

0.64

efe

0.63

oston

0.63

ophys

0.63

obook

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.