INDEX

Explanations

references to influential models or benchmarks in a given field or context

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

inspiration, emulation, or example

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

rippa

-0.45

TagMode

-0.45

IntentFilter

-0.40

contentLoaded

-0.40

 defaultstate

-0.39

OrWhiteSpace

-0.39

ipients

-0.39

ActionCreators

-0.38

 pasada

-0.38

boru

-0.38

POSITIVE LOGITS

 Vorbild

0.42

Tembelea

0.39

 imit

0.38

 Wassers

0.38

模

0.37

 beeinf

0.37

Tikang

0.37

 newBuilder

0.37

 inspire

0.36

 nahilalakip

0.36

Activations Density 0.049%

references to influential models or benchmarks in a given field or context

inspiration, emulation, or example

No Comments

No Known Activations