INDEX

Explanations

mentions of social media followers

oai_token-act-pair · gpt-3.5-turbo

references to social media followers

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

 Genocide

-0.73

Rim

-0.72

ces

-0.72

ced

-0.70

 Prosecutor

-0.67

RAG

-0.66

 circumstance

-0.66

Kob

-0.65

 Scotia

-0.63

 Stars

-0.63

POSITIVE LOGITS

hip

1.28

 followers

1.04

hips

0.97

 follower

0.85

wagon

0.83

lia

0.81

adelphia

0.76

 wagon

0.76

ievers

0.76

azel

0.75

Activations Density 0.012%

mentions of social media followers

references to social media followers

No Comments

No Known Activations