INDEX

Explanations

discourse markers and conjunctions that denote relationships or transitions in text

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

conjunctions and causal links

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 <<<<<<<<<<<<<<

-0.48

 nakalista

-0.48

bkz

-0.48

 gyhoeddwyd

-0.47

 Италијани

-0.45

rola

-0.43

 Meksiku

-0.41

ElementException

-0.41

 zijne

-0.41

platin

-0.40

POSITIVE LOGITS

INCREF

0.63

Šaltiniai

0.50

 Deshalb

0.49

 Çünkü

0.49

RTHOOK

0.49

 Ведь

0.48

GIVEREF

0.45

brigens

0.45

omiast

0.45

GraphicsUnit

0.44

Activations Density 0.004%

discourse markers and conjunctions that denote relationships or transitions in text

conjunctions and causal links

No Comments

No Known Activations