INDEX

Explanations

phrases related to a specific location named "Cal"

oai_token-act-pair · gpt-3.5-turbo

references to a specific location or entity associated with "Cal."

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

ĸļ

-0.89

erness

-0.81

è¦ļéĨĴ

-0.79

ãģį

-0.76

hift

-0.73

kaya

-0.73

ovych

-0.71

 EntityItem

-0.69

DragonMagazine

-0.67

-+

-0.67

POSITIVE LOGITS

endars

1.06

ibr

1.05

iber

0.96

cium

0.93

orie

0.92

iphate

0.90

vin

0.89

houn

0.89

ient

0.89

gary

0.86

Activations Density 0.010%

phrases related to a specific location named "Cal"

references to a specific location or entity associated with "Cal."

No Comments

No Known Activations