INDEX

Explanations

phrases indicating suggestion or implication

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

peli

-0.46

ucksack

-0.46

 chops

-0.45

-0.44

ümü

-0.44

 עלי

-0.43

verständlich

-0.43

bump

-0.42

 strani

-0.42

Exclusive

-0.42

POSITIVE LOGITS

rungsseite

1.01

kloped

0.99

 nahilalakip

0.99

 propOrder

0.96

 CreateTagHelper

0.94

NameInMap

0.93

SpringBootTest

0.91

Билгалдахарш

0.90

GEBURTSDATUM

0.88

EDEFAULT

0.85

Activations Density 0.167%

phrases indicating suggestion or implication

No Comments

No Known Activations