INDEX

Explanations

mentions of professions or job titles

oai_token-act-pair · gpt-3.5-turbo

references to various professional roles or occupations

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

co

-0.69

ston

-0.67

CO

-0.67

TI

-0.67

ces

-0.66

 Territories

-0.65

TON

-0.64

Anything

-0.62

VI

-0.62

tein

-0.61

POSITIVE LOGITS

poons

0.91

paces

0.90

 encount

0.86

heet

0.85

 challeng

0.84

 sugg

0.84

uggest

0.81

 allege

0.81

ioned

0.80

 learned

0.79

Activations Density 0.207%

mentions of professions or job titles

references to various professional roles or occupations

No Comments

No Known Activations