INDEX

Explanations

references to sensitive or controversial topics/issues including surveillance, torture, protests, political decisions, and governmental actions

oai_token-act-pair · gpt-3.5-turbo

references to sensitive information and actions associated with accountability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

maxwell

-0.87

ãĤ¦ãĤ¹

-0.84

 teasp

-0.79

Appearance

-0.76

 evolves

-0.72

Interstitial

-0.70

ipel

-0.68

Present

-0.68

-->

-0.68

çĶ°

-0.67

POSITIVE LOGITS

 alleged

0.79

 Confederate

0.76

 disputed

0.76

 suspected

0.76

 controversial

0.74

 Uganda

0.72

 Guantanamo

0.72

 Libya

0.71

 Guant

0.71

 Nazi

0.70

Activations Density 0.471%

references to sensitive or controversial topics/issues including surveillance, torture, protests, political decisions, and governmental actions

references to sensitive information and actions associated with accountability

No Comments

No Known Activations