INDEX

Explanations

terms related to relationships and construction

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

Follows "Q:" or indicates the beginning of a question

np_acts-logits-general · gemini-2.0-flash

perhaps differentiation or captured

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

מעט

-0.50

幾乎

-0.47

基本的に

-0.42

几乎

-0.42

стоян

-0.41

 virtually

-0.41

PositiveButton

-0.41

 основном

-0.40

 vět

-0.39

 téměř

-0.39

POSITIVE LOGITS

 perhaps

1.43

 maybe

1.42

perhaps

1.32

 somehow

1.22

 måske

1.22

maybe

1.21

 perchance

1.20

 vielleicht

1.19

 Perhaps

1.17

 Maybe

1.17

Activations Density 0.747%

terms related to relationships and construction

Follows "Q:" or indicates the beginning of a question

perhaps differentiation or captured

No Comments

No Known Activations