INDEX

Explanations

mentions of positions or roles held by individuals within organizations

oai_token-act-pair · gpt-3.5-turbo

references to individuals in leadership positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

SPONSORED

-0.86

holes

-0.78

LOAD

-0.70

loads

-0.69

ictions

-0.68

rums

-0.68

 followed

-0.68

 alone

-0.68

load

-0.67

fulness

-0.67

POSITIVE LOGITS

 Association

0.99

 International

0.90

 Institute

0.88

 aforementioned

0.88

 National

0.88

 Philippine

0.86

 Knights

0.86

 Federation

0.85

 infamous

0.85

 Anti

0.85

Activations Density 0.168%

mentions of positions or roles held by individuals within organizations

references to individuals in leadership positions

No Comments

No Known Activations