INDEX

Explanations

terms related to legal or regulatory contexts

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

preceding nouns related to formal processes

np_acts-logits-general · gemini-2.0-flash

used strategies, former Chief, output terminals

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

TestingModule

-0.71

 vœux

-0.65

 bienfaits

-0.64

 épaules

-0.63

 modalités

-0.60

 attentes

-0.59

 autorités

-0.58

-0.57

 conséquences

-0.57

 inconvénients

-0.57

POSITIVE LOGITS

 histories

0.91

 models

0.87

 ones

0.86

ങ്ങള

0.86

 programs

0.85

ographies

0.84

 complexes

0.84

 versions

0.83

ങ്ങൾ

0.82

 statuses

0.80

Activations Density 2.887%

terms related to legal or regulatory contexts

preceding nouns related to formal processes

used strategies, former Chief, output terminals

No Comments

No Known Activations