INDEX

Explanations

references to specific individuals or organizations

oai_token-act-pair · gpt-3.5-turbo

recurring conjunctions and prepositions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Niet

-0.67

 Highlander

-0.62

 Bernstein

-0.61

 Begins

-0.60

 Petraeus

-0.59

phrine

-0.57

 targ

-0.56

 Luthor

-0.56

 Oval

-0.56

 Kahn

-0.54

POSITIVE LOGITS

zona

0.65

soDeliveryDate

0.61

TPPStreamerBot

0.60

avis

0.60

ï¸ı

0.60

^{

0.58

_{

0.57

arge

0.57

xit

0.56

Ô

0.56

Activations Density 0.585%

references to specific individuals or organizations

recurring conjunctions and prepositions

No Comments

No Known Activations