INDEX

Explanations

connections to metaphors and idiomatic expressions

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

idiomatic expressions

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.31

 stay

-0.27

 brukt

-0.26

 used

-0.25

 give

-0.25

 unit

-0.24

 plan

-0.24

ph

-0.24

 licensed

-0.24

en

-0.24

POSITIVE LOGITS

 AssemblyCompany

0.84

rungsseite

0.79

 ویکی‌پدی

0.77

󠁢

0.75

 चीज़ों

0.75

最快更新

0.75

StoryboardSegue

0.72

 typelib

0.72

+#+#

0.71

IndentedString

0.71

Activations Density 0.057%

connections to metaphors and idiomatic expressions

idiomatic expressions

No Comments

No Known Activations