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

otal

-0.76

poke

-0.73

ction

-0.71

valued

-0.65

anza

-0.62

rent

-0.62

ctive

-0.60

yss

-0.60

heimer

-0.60

verse

-0.59

POSITIVE LOGITS

ul

0.74

 attackers

0.66

 corrid

0.63

 safest

0.60

ector

0.59

hel

0.59

 bombers

0.59

 attacker

0.58

 Assassins

0.57

Gut

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.