INDEX

Explanations

elements related to actions and interactions among characters

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

they appear to emerge

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

 transfieras

-0.44

ßerdem

-0.42

FXMLLoader

-0.42

middelen

-0.41

عرِّف

-0.39

 ressources

-0.39

gbar

-0.39

Ric

-0.39

Succeed

-0.39

ONA

-0.38

POSITIVE LOGITS

JspWriter

0.49

 विश्वसनीयता

0.47

piram

0.45

MessageState

0.41

tranno

0.41

Décès

0.40

ArrowToggle

0.38



0.38

SourceChecksum

0.38

AnchorTagHelper

0.37

Activations Density 0.040%

elements related to actions and interactions among characters

they appear to emerge

No Comments

No Known Activations