INDEX

Explanations

specific names and terms associated with academic and professional credentials or affiliations

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

 Hake

-0.55

 Waco

-0.49

iya

-0.47

genic

-0.47

 Sangat

-0.47

 Osun

-0.47

jarati

-0.46

 Texas

-0.46

HttpFoundation

-0.46

 Joni

-0.46

POSITIVE LOGITS

 Bulgarian

1.10

 Bulgaria

1.06

Bulgaria

0.92

Bulg

0.91

NewUrlParser

0.85

 Sofia

0.84

 Bulgar

0.81

Sofia

0.79

 Bulg

0.75

 насељу

0.73

Activations Density 0.047%

specific names and terms associated with academic and professional credentials or affiliations

No Comments

No Known Activations