INDEX

Explanations

questions or phrases that inquire about the nature or perception of things

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

Follows the word "what."

np_acts-logits-general · gemini-2.0-flash

what is/was

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

UnsafeEnabled

-0.93

RegressionTest

-0.81

 AssemblyTitle

-0.81

enterOuterAlt

-0.78

elemField

-0.70

IContainer

-0.68

 فريبيس

-0.68

IntoConstraints

-0.66

tagHelperRunner

-0.64

 ProtoMessage

-0.64

POSITIVE LOGITS

 essentially

0.82

 arguably

0.81

 amounts

0.76

 otherwise

0.68

 amounted

0.65

 admittedly

0.65

essentially

0.63

 termed

0.63

 probably

0.62

 surely

0.62

Activations Density 0.100%

questions or phrases that inquire about the nature or perception of things

Follows the word "what."

what is/was

No Comments

No Known Activations