INDEX

Explanations

evidence or indicators related to judgement and perception

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

deduction or assessment

np_acts-logits-general · gemini-2.0-flash

inferring from evidence

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

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

WebMethod

-0.46

こそ

-0.46

uolo

-0.45

μία

-0.44

怎么能

-0.44

nowu

-0.43

AddComponent

-0.42

소

-0.42

kháu

-0.42

 ideas

-0.42

POSITIVE LOGITS

 judging

1.56

Judging

1.45

 Judging

1.44

 based

1.24

 from

1.21

 Based

1.20

Based

1.09

 looking

1.07

 BASED

1.07

 berdasarkan

1.03

Activations Density 0.533%

evidence or indicators related to judgement and perception

deduction or assessment

inferring from evidence

No Comments

No Known Activations