INDEX

Explanations

references to URLs and options in a programming or API context

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

technical identifiers and measurements

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

Климат

-0.59

 EconPapers

-0.54

pexpr

-0.48

+#+#

-0.47

 pinulongan

-0.45

felves

-0.44

QMetaType

-0.43

 Opti

-0.42

ByVersion

-0.42

loven

-0.41

POSITIVE LOGITS

 that

0.49

 where

0.45

mà

0.42

 जहां

0.42

 deployed

0.41

IVEREF

0.41

 which

0.40

corazón

0.40

wnież

0.39

 involved

0.38

Activations Density 0.042%

references to URLs and options in a programming or API context

technical identifiers and measurements

No Comments

No Known Activations