INDEX

Explanations

references to specific events, achievements, or milestones

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

quantifiers and events

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

parsedMessage

-0.77

 uſed

-0.58

 ſtand

-0.57

ThroughAttribute

-0.56

GOTREF

-0.56

 Reſ

-0.56

 BorderSide

-0.55

 ſind

-0.54

(!__

-0.53

новништво

-0.52

POSITIVE LOGITS

 yearly

0.38

 annual

0.35

țională

0.33

 autorytatywna

0.32

 <=",

0.31

举办

0.31

StoryboardSegue

0.30

 COLLECT

0.28

 Administrativna

0.28

SequentialGroup

0.28

Activations Density 0.101%

references to specific events, achievements, or milestones

quantifiers and events

No Comments

No Known Activations