INDEX

Explanations

names, specifically focusing on the name "Mika."

oai_token-act-pair · gpt-3.5-turbo

names that contain the substring "ika"

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

ensible

-0.82

 corrid

-0.78

lying

-0.76

ingham

-0.76

 glim

-0.75

gob

-0.74

iage

-0.74

ĻĤ

-0.73

dit

-0.72

auld

-0.70

POSITIVE LOGITS

ika

0.90

emia

0.85

Suk

0.83

ota

0.81

Å¡

0.80

uchi

0.80

Lok

0.78

ppa

0.78

qt

0.77

Mae

0.77

Activations Density 0.016%

names, specifically focusing on the name "Mika."

names that contain the substring "ika"

No Comments

No Known Activations