INDEX

Explanations

keywords related to programming extensions or modifications

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

extension and additions

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

Tembelea

-0.59

__;

-0.52

 <>",

-0.51

 beginnetje

-0.50

'{@

-0.48

 propOrder

-0.47

LogFactory

-0.47

GenerationType

-0.46

IANGLES

-0.46

wpi

-0.45

POSITIVE LOGITS

 Erweiterung

0.64

Extended

0.63

 supplé

0.63

 extras

0.62

 Extras

0.62

 rozší

0.61

extras

0.61

 addition

0.61

 tambahan

0.60

 extension

0.60

Activations Density 0.068%

keywords related to programming extensions or modifications

extension and additions

No Comments

No Known Activations