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

yip

-0.75

quet

-0.72

xe

-0.67

agram

-0.67

eret

-0.67

':

-0.66

eline

-0.66

chron

-0.66

iami

-0.66

furt

-0.66

POSITIVE LOGITS

 marginal

0.86

 decriminal

0.65

 reinforcing

0.63

 moderation

0.62

 reinforce

0.61

 ridicule

0.60

 misconceptions

0.60

 Dism

0.60

 corro

0.60

 Guys

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.