INDEX

Explanations

positions or titles related to editorial and managerial roles

oai_token-act-pair · gpt-3.5-turbo

references to editorial or leadership roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 whim

-0.73

 chaotic

-0.69

ulus

-0.65

 decency

-0.65

 stranger

-0.65

 suscept

-0.64

namese

-0.64

 peril

-0.64

 catast

-0.64

rums

-0.64

POSITIVE LOGITS

 overseeing

0.75

 Dean

0.73

arsity

0.71

bernatorial

0.70

 Assist

0.69

bard

0.68

ective

0.68

 adjunct

0.68

RECT

0.67

 Editorial

0.64

Activations Density 0.232%

positions or titles related to editorial and managerial roles

references to editorial or leadership roles

No Comments

No Known Activations