INDEX

Explanations

sequences of protected attributes or variables in a code context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

antMatchers

-0.71

 jLabel

-0.62

UVWXYZ

-0.59

":"+

-0.56

міст

-0.56

iama

-0.54

='"+

-0.53

':'

-0.53

 Clarke

-0.53

('--

-0.53

POSITIVE LOGITS

RepeatedField

0.78



0.75

تقاوى

0.69

Vidite

0.65

RRect

0.65

 defaultstate

0.65

 Reſ

0.64

 समीक्षाओं

0.63

 herself

0.63

ſelves

0.62

Activations Density 0.054%

sequences of protected attributes or variables in a code context

No Comments

No Known Activations