INDEX

Explanations

terms related to Markov equivalence and its characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Lopes

-0.36

junto

-0.35

ติด

-0.35

 Robinson

-0.34

vel

-0.34

Mo

-0.33

 Convenient

-0.33

son

-0.33

concaten

-0.33

Ina

-0.33

POSITIVE LOGITS

Personensuche

0.58

 صوتيه

0.57

PullParser

0.54

QMetaType

0.54

meral

0.52

majánló

0.52

iſche

0.50

ſicht

0.50

ésultats

0.49

 createSprite

0.49

Activations Density 0.775%

terms related to Markov equivalence and its characteristics

No Comments

No Known Activations