INDEX

Explanations

instances of function and method calls related to variable and object manipulation in code

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

directory paths and file names

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

 BrowserModule

-0.79

 snippetHide

-0.70

 faſt

-0.57

 ✭✭

-0.52

Geplaatst

-0.51

+#+

-0.49

awtextra

-0.49

oredCriteria

-0.47

 ſch

-0.47

ſel

-0.47

POSITIVE LOGITS

GraphicsUnit

0.47

 saurait

0.44

GeoNames

0.37

classnames

0.37

getColumnIndex

0.36

 McAllister

0.35

 assuming

0.35

 meille

0.34

 senescence

0.34

 senjata

0.33

Activations Density 0.039%

instances of function and method calls related to variable and object manipulation in code

directory paths and file names

No Comments

No Known Activations