INDEX

Explanations

high-ranking government officials and officials in government positions

oai_token-act-pair · gpt-3.5-turbo

mentions of high-ranking officials or government 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

crew

-0.95

skill

-0.81

 needle

-0.77

circle

-0.74

 curve

-0.71

eat

-0.70

pless

-0.69

 panc

-0.69

MpServer

-0.67

xs

-0.67

POSITIVE LOGITS

Tim

0.88

 Manager

0.86

 Brett

0.85

 Shaun

0.84

 Ambassador

0.83

 Adviser

0.83

 Lyndon

0.81

Won

0.81

 Mitch

0.80

 Andrew

0.80

Activations Density 0.102%

high-ranking government officials and officials in government positions

mentions of high-ranking officials or government positions

No Comments

No Known Activations