INDEX

Explanations

sequences of whitespace or indentation in code

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

user code interactions

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

 ſta

-0.66

 Reſ

-0.56

 raiſ

-0.56

 Houſe

-0.54

 houſe

-0.52

 itſelf

-0.51

ſelves

-0.51

 Perſ

-0.50

 étoit

-0.49

 Diſ

-0.48

POSITIVE LOGITS

DockStyle

0.58

.*")]

0.55

ScopeManager

0.49

ंदीखरीदारी

0.48

ExtendWith

0.48

 ExecuteAsync

0.47

ftagPool

0.47

webElementXpaths

0.45

}));

0.44

fjspx

0.44

Activations Density 0.002%

sequences of whitespace or indentation in code

user code interactions

No Comments

No Known Activations