INDEX

Explanations

terms that indicate speed or quickness

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

quickness

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

outState

-0.44

 finnas

-0.43

enumi

-0.43

lapsingToolbar

-0.43

.*")]

-0.40

DataBinding

-0.39

MockBean

-0.38

 onCreateView

-0.38

minecraftforge

-0.38

endpush

-0.38

POSITIVE LOGITS

 quick

1.09

quick

1.06

 rápido

1.06

 cepat

1.06

 quickly

1.03

quickly

1.02

Quick

1.00

 rapid

0.98

 rápida

0.98

 fast

0.96

Activations Density 0.040%

terms that indicate speed or quickness

quickness

No Comments

No Known Activations