INDEX

Explanations

phrases that indicate uncertainty or speculation regarding opinions and events

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

expressing possibility or speculation

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

AddTagHelper

-0.65

awtextra

-0.58

webElementXpaths

-0.57

 betweenstory

-0.56

expandindo

-0.56

tagHelperRunner

-0.56

 ProtoMessage

-0.54

 Савезне

-0.51

 CreateTagHelper

-0.51

 InputDecoration

-0.51

POSITIVE LOGITS

 madness

0.43

badlogic

0.40

ništvo

0.39

 kona

0.37

 overflow

0.37

 damn

0.37

EDM

0.37

Reign

0.37

 generation

0.36

 Aristotle

0.36

Activations Density 0.022%

phrases that indicate uncertainty or speculation regarding opinions and events

expressing possibility or speculation

No Comments

No Known Activations