INDEX

Explanations

terms related to government agencies

oai_token-act-pair · gpt-3.5-turbo

mentions of an agency or organization

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

 Shields

-0.74

Tea

-0.69

 Corbyn

-0.68

 Wast

-0.67

âķĲâķĲ

-0.67

 Bened

-0.67

 Lect

-0.66

Klu

-0.65

 Ellison

-0.64

ix

-0.64

POSITIVE LOGITS

 agency

1.22

 agencies

1.01

agency

0.86

 Agency

0.81

 agent

0.80

encies

0.79

agents

0.79

acco

0.78

 contractor

0.77

 therap

0.77

Activations Density 0.012%

terms related to government agencies

mentions of an agency or organization

No Comments

No Known Activations