INDEX

Explanations

phrases related to negative social behaviors such as harassment, threats, and discrimination

oai_token-act-pair · gpt-3.5-turbo

phrases and concepts related to abuse and social issues

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

çīĪ

-0.72

ufact

-0.69

arten

-0.66

ppa

-0.62

illac

-0.61

ainted

-0.59

puter

-0.59

hesion

-0.57

ahon

-0.55

ixel

-0.55

POSITIVE LOGITS

 imprisonment

0.73

 retribution

0.70

 bullying

0.70

 persecution

0.70

 degradation

0.68

 threats

0.67

 boredom

0.67

 consequ

0.67

 exploitation

0.67

 harassment

0.65

Activations Density 0.486%

phrases related to negative social behaviors such as harassment, threats, and discrimination

phrases and concepts related to abuse and social issues

No Comments

No Known Activations