INDEX

Explanations

mention of political events and figures

oai_token-act-pair · gpt-3.5-turbo

references to political events and actions

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

↵Âł

-0.89

 whilst

-0.72

 Whilst

-0.68

...)

-0.68

ÂłÂł

-0.66

 Âł Âł Âł Âł

-0.65

;)

-0.65

!!

-0.65

 english

-0.65

âĢ¦)

-0.64

POSITIVE LOGITS

 Fiorina

0.76

 Cohn

0.70

 Buffett

0.69

 Southwest

0.69

 POLITICO

0.69

 DeVos

0.67

 Gorsuch

0.66

 bipartisan

0.65

©¶æ

0.64

 Sanders

0.64

Activations Density 2.921%

mention of political events and figures

references to political events and actions

No Comments

No Known Activations