INDEX

Explanations

conjunctions and phrases indicating multiple actions or items

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

Word followed by imperative verb

np_acts-logits-general · gemini-2.0-flash

and let

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

TestingModule

-0.67

 linkovi

-0.65

 đã

-0.61

NameInMap

-0.61

 mergeFrom

-0.60

+#+#

-0.60

 Мексичка

-0.59

Personensuche

-0.59

HideFlags

-0.59

OGND

-0.58

POSITIVE LOGITS

let

1.19

don

1.09

 watch

1.02

 remember

0.96

 keep

0.93

try

0.93

 please

0.93

 jangan

0.92

 enjoy

0.90

ask

0.89

Activations Density 0.140%

conjunctions and phrases indicating multiple actions or items

Word followed by imperative verb

and let

No Comments

No Known Activations