INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

fw

-0.77

Thumbnail

-0.77

Translation

-0.73

 pinch

-0.70

 dispatcher

-0.70

PF

-0.68

 commander

-0.66

pg

-0.66

Lt

-0.66

TN

-0.64

POSITIVE LOGITS

ophon

0.80

ocard

0.77

gans

0.77

luaj

0.74

sonian

0.70

ament

0.69

ourke

0.68

heses

0.67

amation

0.66

rals

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.