INDEX

Explanations

references to events in time, particularly focusing on prior occurrences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.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

 ويكيميديا

-0.65

evap

-0.63

zeň

-0.59

kháu

-0.57

arova

-0.56

Whisper

-0.55

 sonno

-0.55

odle

-0.55

Kni

-0.54

 rapides

-0.54

POSITIVE LOGITS

 earlier

0.82

 Earlier

0.81

Earlier

0.81

Ранее

0.72

 previously

0.70

 eerder

0.65

earlier

0.64

CppMethod

0.63

AndEndTag

0.63

 sebelumnya

0.61

Activations Density 0.025%

references to events in time, particularly focusing on prior occurrences

No Comments

No Known Activations