INDEX

Explanations

terms related to rankings and arrangements of materials based on their properties

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

yttö

-0.52

Rüyada

-0.51

 supplémentaires

-0.51

 الحره

-0.46

unref

-0.46

Hauptartikel

-0.46

assertNot

-0.45

Enllaces

-0.44

 ब्रेकडाउन

-0.44

estanden

-0.43

POSITIVE LOGITS

 descending

1.18

 downwards

1.15

 backwards

1.12

 ascending

1.12

 decreasing

1.11

 clockwise

1.08

 downward

1.07

 backward

0.98

decreasing

0.96

 progressively

0.96

Activations Density 0.318%

terms related to rankings and arrangements of materials based on their properties

No Comments

No Known Activations