INDEX

Explanations

references to specific dates and events in time

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

year mentions

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 Italijani

-0.37

énario

-0.37

IMPLEMENT

-0.36

 estructural

-0.35

 implemented

-0.34

 layak

-0.34

 weder

-0.34

 niestety

-0.34

 niezbęd

-0.33

 توانند

-0.33

POSITIVE LOGITS

 snippetHide

0.62

tagHelperRunner

0.60

 wireType

0.56

TagMode

0.55

quelize

0.54

richTextPanel

0.53

ValueStyle

0.51

 photographing

0.49

 touring

0.48

 tropo

0.47

Activations Density 0.042%

references to specific dates and events in time

year mentions

No Comments

No Known Activations