INDEX

Explanations

phrases indicating value or worth

oai_token-act-pair · gpt-3.5-turbo

phrases related to the value or worth of something in relation to effort or cost

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

apons

-0.71

ADRA

-0.70

ARDS

-0.70

OHN

-0.67

PLA

-0.65

unn

-0.61

hent

-0.60

ardo

-0.59

 sclerosis

-0.59

yrinth

-0.59

POSITIVE LOGITS

 dearly

1.00

$$$$

0.85

 gamble

0.82

 hassle

0.77

$$

0.77

 peanuts

0.75

$$

0.74

 upfront

0.74

inar

0.73

 upkeep

0.73

Activations Density 0.124%

phrases indicating value or worth

phrases related to the value or worth of something in relation to effort or cost

No Comments

No Known Activations