INDEX

Explanations

the article "a" indicating its frequent use as a determiner

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 itſelf

-0.91

 myſelf

-0.91

 ſhe

-0.80

 Jefus

-0.78

 whoſe

-0.77

 themſelves

-0.76

 Theſe

-0.75

 himſelf

-0.75

 becauſe

-0.74

 houſe

-0.74

POSITIVE LOGITS

CloseOperation

0.84

For

0.82

FOR

0.81

FOR

0.80

for

0.79

For

0.76

for

0.71

 für

0.67

 fürs

0.66

ForValue

0.62

Activations Density 0.103%

the article "a" indicating its frequent use as a determiner

No Comments

No Known Activations