INDEX

Explanations

keywords related to weighing the advantages and disadvantages of a situation or decision

oai_token-act-pair · gpt-3.5-turbo

references to advantages and disadvantages in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 faiths

-0.72

gments

-0.70

 Prophe

-0.69

 Galile

-0.66

 pige

-0.63

 prisoners

-0.62

ĭ

-0.62

 journeys

-0.61

 Bird

-0.61

 Sutherland

-0.61

POSITIVE LOGITS

creen

1.33

ervative

1.14

ervatives

1.13

heet

1.09

aic

1.03

uits

1.03

linger

0.96

igl

0.94

cale

0.94

peed

0.93

Activations Density 0.199%

keywords related to weighing the advantages and disadvantages of a situation or decision

references to advantages and disadvantages in various contexts

No Comments

No Known Activations