INDEX

Explanations

key terms and metrics related to events, actions, and conditions

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

words before articles or prepositions

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

⇄

-0.36

ろ

-0.35

 enda

-0.35

ymce

-0.35

 McQu

-0.34

ConstraintMaker

-0.33

dique

-0.33

nalité

-0.33

les

-0.33

empel

-0.32

POSITIVE LOGITS

 виправивши

0.68

DoubleQuotes

0.66

 both

0.66

both

0.64

OGND

0.63

出版年

0.61

BOTH

0.60

 entrambi

0.60

matchCondition

0.59

 båda

0.59

Activations Density 0.053%

key terms and metrics related to events, actions, and conditions

words before articles or prepositions

No Comments

No Known Activations