INDEX

Explanations

references to arrays or list structures in programming contexts

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

bracketed code segments, indexing

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

 betweenstory

-0.76

Personendaten

-0.57

FunctionFlags

-0.56



-0.54

transQ

-0.54

下载附件

-0.52

点此举报

-0.51

layoutControl

-0.51

EndContext

-0.49

ransition

-0.47

POSITIVE LOGITS

 famí

0.38

 Jerusalén

0.36

MigrationBuilder

0.35

 المعيارى

0.34

 Keluarga

0.32

 sorella

0.32

Referències

0.31

segno

0.30

 suaminya

0.29

 география

0.29

Activations Density 0.003%

references to arrays or list structures in programming contexts

bracketed code segments, indexing

No Comments

No Known Activations