INDEX

Explanations

code structures and interactions in programming contexts

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

byte accesses and indices

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

ltk

-0.42

 SDLK

-0.37

 XCTest

-0.36

 Einwilligung

-0.36

ísimas

-0.35

emlrt

-0.34

riwal

-0.34

uidado

-0.33

;"></

-0.33

 Polak

-0.32

POSITIVE LOGITS

AnchorStyles

0.61

 defaultstate

0.52

postIndex

0.51

DockStyle

0.49

 informée

0.48

AccessFile

0.48

Демографія

0.45

tvb

0.45

 cellspacing

0.44

adpleegd

0.44

Activations Density 0.192%

code structures and interactions in programming contexts

byte accesses and indices

No Comments

No Known Activations