INDEX

Explanations

references to the action of closing in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

CppMethod

-0.55

Viki

-0.52

menuStrip

-0.50

дивиду

-0.47

 Voit

-0.46

 Viking

-0.46

 Disney

-0.45

 Ivar

-0.45

 Vector

-0.45

 Effort

-0.44

POSITIVE LOGITS

Close

1.22

 CLOSE

1.20

close

1.20

 Close

1.14

 close

1.12

CLOSE

1.08

 CLOS

0.96

 closes

0.94

 Closes

0.94

 Closing

0.93

Activations Density 0.101%

references to the action of closing in various contexts

No Comments

No Known Activations