INDEX

Explanations

technical terms and specifics related to scientific structures and processes

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

specific particular types

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

Tikang

-0.50

verifyException

-0.49

titleMargin

-0.45

Vidite

-0.45

encodeWith

-0.44

 tours

-0.44

Chham

-0.43

stande

-0.41

 HttpNotFound

-0.41

piram

-0.40

POSITIVE LOGITS

 específico

0.55

 particular

0.54

 determinado

0.53

 específica

0.52

particular

0.52

 determinada

0.50

 okreś

0.47

Particular

0.46

 tertentu

0.46

 Particular

0.46

Activations Density 0.220%

technical terms and specifics related to scientific structures and processes

specific particular types

No Comments

No Known Activations