INDEX

Explanations

programming syntax and structure related to code import statements

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

java and other code

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

 createSprite

-0.53

PutMapping

-0.41

flashdata

-0.41

汰

-0.40

FDRE

-0.39

❉

-0.36

borderSide

-0.36

FXMLLoader

-0.35

солю

-0.35

postIndex

-0.34

POSITIVE LOGITS

 ujednoznacz

0.65

 AssemblyCulture

0.63

CppMethod

0.63

 propTypes

0.57

sizeCache

0.54

propTypes

0.53

 nakalista

0.52

 PropTypes

0.48

 bailarina

0.46

 vagu

0.46

Activations Density 0.002%

programming syntax and structure related to code import statements

java and other code

No Comments

No Known Activations