INDEX

Explanations

concepts related to spatial location and environment

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

preceding locations ("stage", "clinic", "membrane," etc.)

np_acts-logits-general · gemini-2.0-flash

membrane expression

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

StoreMessageInfo

-0.46

出版年

-0.46

 createState

-0.44

providedIn

-0.43

verwijspagina

-0.41

 localObject

-0.40

 BoxDecoration

-0.40

 Holanda

-0.39

describe

-0.39

 Verſ

-0.39

POSITIVE LOGITS

 dwellers

0.63

borne

0.51

上の

0.51

seitige

0.51

bound

0.51

での

0.50

上的

0.49

 ivelany

0.49

ward

0.49

side

0.47

Activations Density 0.760%

concepts related to spatial location and environment

preceding locations ("stage", "clinic", "membrane," etc.)

membrane expression

No Comments

No Known Activations