INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Learns

-0.69

ithe

-0.68

pav

-0.64

 Polar

-0.63

Pav

-0.62

 Atari

-0.61

thood

-0.60

//[

-0.59

 drive

-0.58

 Aram

-0.58

POSITIVE LOGITS

roxy

0.87

otal

0.80

 Thomson

0.75

ammers

0.71

aque

0.70

heses

0.69

otos

0.68

ricks

0.68

©¶æ¥µ

0.67

raviolet

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.