INDEX

Explanations

words related to legal actions and political decisions

oai_token-act-pair · gpt-3.5-turbo

references to legislative or judicial actions and decisions

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

Live

-0.66

 Named

-0.63

Weather

-0.58

Cho

-0.57

adeon

-0.57

etimes

-0.56

verbs

-0.54

hem

-0.54

src

-0.53

parts

-0.53

POSITIVE LOGITS

fulness

0.68

 debacle

0.67

 inaccur

0.65

 prompted

0.62

 embold

0.62

 anonymously

0.59

 onslaught

0.59

HMS

0.59

 conclude

0.56

 spree

0.56

Activations Density 0.429%

words related to legal actions and political decisions

references to legislative or judicial actions and decisions

No Comments

No Known Activations