INDEX

Explanations

mentions of distinctive groups or categories within a larger population

oai_token-act-pair · gpt-3.5-turbo

references to specific groups of people and their roles or professions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

uden

-0.99

endi

-0.91

 Lumpur

-0.89

Ô

-0.86

phas

-0.85

onite

-0.84

hesda

-0.83

enth

-0.82

oria

-0.79

uesday

-0.76

POSITIVE LOGITS

units

0.71

SHIP

0.68

TTC

0.68

 minions

0.67

 cohorts

0.67

 dummy

0.67

 careg

0.67

 SERV

0.67

 Bagg

0.66

 helper

0.66

Activations Density 0.032%

mentions of distinctive groups or categories within a larger population

references to specific groups of people and their roles or professions

No Comments

No Known Activations