INDEX

Explanations

phrases indicating connections, references, or associations between entities

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

lists connected by and

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

 initComponents

-0.39

 Fast

-0.35

ウザ

-0.35

setAge

-0.34

 Tension

-0.33

ьажоргаш

-0.33

Age

-0.32

 שוליים

-0.32

jud

-0.32

anja

-0.32

POSITIVE LOGITS

appartamento

0.59

BufferException

0.58

 surla

0.57

 &___

0.56

 nahilalakip

0.52

SBATCH

0.52

balleur

0.51

 للاسماء

0.48

Personensuche

0.48

 resourceCulture

0.48

Activations Density 0.081%

phrases indicating connections, references, or associations between entities

lists connected by and

No Comments

No Known Activations