INDEX

Explanations

positions or titles within organizations

oai_token-act-pair · gpt-3.5-turbo

titles and roles associated with leadership positions

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

 routine

-0.69

IRO

-0.68

 save

-0.64

 Colleges

-0.61

AAF

-0.60

 breeze

-0.60

 limitless

-0.59

 audition

-0.58

 whim

-0.57

 timetable

-0.57

POSITIVE LOGITS

itatively

0.91

 emer

0.88

uably

0.87

director

0.84

 Managing

0.78

manager

0.75

 director

0.73

 Emer

0.72

 Paste

0.71

buquerque

0.67

Activations Density 0.132%

positions or titles within organizations

titles and roles associated with leadership positions

No Comments

No Known Activations