INDEX

Explanations

concepts related to thinking and understanding philosophical questions

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

Followed by infinitive form of a verb ("to [verb]")

np_acts-logits-general · gemini-2.0-flash

to + verb infinitive

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 jsPsych

-0.82

CloseOperation

-0.71

 ProtoMessage

-0.65

etcode

-0.64

 TextInputType

-0.63

 wireType

-0.63

 transférez

-0.63

GenerationType

-0.62

DotNetBar

-0.60

etc

-0.60

POSITIVE LOGITS

 damned

0.78

 damn

0.71

 goddamn

0.71

 fucking

0.66

 darn

0.63

 сами

0.61

 actually

0.59

 самого

0.57

 precisamente

0.57

 freakin

0.56

Activations Density 0.609%

concepts related to thinking and understanding philosophical questions

Followed by infinitive form of a verb ("to [verb]")

to + verb infinitive

No Comments

No Known Activations