INDEX

Explanations

complex relationships and elements within systems

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

preceding prepositions

np_acts-logits-general · gemini-2.0-flash

state or characteristic

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

LookAnd

-0.79

 AssemblyTitle

-0.76

ViewFeatures

-0.74

ftagPool

-0.70

windowFixed

-0.68

 snippetHide

-0.66

adaptiveStyles

-0.66

WriteTagHelper

-0.64

 CreateTagHelper

-0.63

GOTREF

-0.61

POSITIVE LOGITS

 amarillas

0.32

mobile

0.31

⑤

0.31

 expect

0.31

getJSONObject

0.31

←

0.30

ACTUAL

0.30

KEN

0.30

bott

0.29

Mike

0.29

Activations Density 2.518%

complex relationships and elements within systems

preceding prepositions

state or characteristic

No Comments

No Known Activations