INDEX

Explanations

names or terms related to legal or military personnel

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to locations and entities, particularly in legal or policing contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Creep

-0.74

 Gaul

-0.74

 Secondary

-0.71

 Pref

-0.68

 Course

-0.66

 Feather

-0.65

 Medic

-0.63

 Magnet

-0.63

 Stream

-0.63

 Pixie

-0.63

POSITIVE LOGITS

atche

0.97

ionage

0.96

ricanes

0.90

ruction

0.90

racuse

0.87

riors

0.86

ricane

0.85

reath

0.85

atan

0.84

acus

0.83

Activations Density 0.039%

names or terms related to legal or military personnel

proper nouns related to locations and entities, particularly in legal or policing contexts

No Comments

No Known Activations