INDEX

Explanations

names of authors or researchers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Monfieur

-0.95

 ―――――

-0.92

 NSCoder

-0.92

 gddr

-0.87

 myſelf

-0.85

 itſelf

-0.84

ReusableCell

-0.84

 Houſe

-0.82

 ویکی‌پدیا

-0.82

AndroidJUnit

-0.81

POSITIVE LOGITS

et

2.88

et

1.30

等人

1.08

etc

1.01

Et

1.00

Et

1.00

等

0.98

등

0.95

etal

0.93

ET

0.86

Activations Density 0.805%

names of authors or researchers

No Comments

No Known Activations