INDEX

Explanations

instances of experiences that are novel or unfamiliar

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

migrationBuilder

-0.64

middels

-0.63

zsche

-0.59

 stále

-0.58

 defaultstate

-0.58

లాలు

-0.58

ņš

-0.58

:]:

-0.57

følgelig

-0.57

 wciąż

-0.56

POSITIVE LOGITS

 auparavant

1.02

ngdoc

0.74

过

0.73

 zuvor

0.69

 EVER

0.68

过的

0.67

過的

0.66

有过

0.66

過

0.65

过

0.64

Activations Density 0.215%

instances of experiences that are novel or unfamiliar

No Comments

No Known Activations