INDEX

Explanations

expressions of personal experience and subjective assessments

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

bootstrapcdn

-0.72

انجليز

-0.65

存于互联网档案馆

-0.65

">—

-0.63

 referenties

-0.60

parsedMessage

-0.60

AsUp

-0.59



-0.58

曖昧さ回避

-0.57

יוחד

-0.57

POSITIVE LOGITS

 ever

1.93

 EVER

1.45

 Ever

1.37

ever

1.34

Ever

1.34

EVER

1.07

 jamais

0.94

 pernah

0.93

 jemals

0.93

 jamás

0.77

Activations Density 0.128%

expressions of personal experience and subjective assessments

No Comments

No Known Activations