INDEX

Explanations

instances of the word "en" in various contexts

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

The French word "en" followed by a noun

np_acts-logits-general · gemini-2.0-flash

en followed by various words

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 Shakspeare

-1.11

 Shaksp

-1.06

 Cæsar

-0.93

 Monfieur

-0.90

 myſelf

-0.89

 gills

-0.89

 itſelf

-0.89

 ་་

-0.89

 Mahomet

-0.87

Efq

-0.87

POSITIVE LOGITS

 form

0.54

 Form

0.46

 base

0.46

cas

0.43

 ocasiones

0.43

 forma

0.42

ing

0.41

 self

0.41

 general

0.40

Activations Density 0.040%

instances of the word "en" in various contexts

The French word "en" followed by a noun

en followed by various words

No Comments

No Known Activations