INDEX

Explanations

references to data structure manipulations and programming-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

RenderAtEndOf

-0.56

UnusedPrivate

-0.45

TypedDataSet

-0.43

AutoresizingMask

-0.38

+#+

-0.38

__);

-0.37

 mergeFrom

-0.37

(((

-0.36

rrggbb

-0.36

*)((

-0.35

POSITIVE LOGITS

contentLoaded

0.55

 laſſen

0.52

 ujednoznacz

0.52

باردا

0.50

 müſſen

0.47

 Waſſer

0.46

iſchen

0.45

 zuſammen

0.43

 IOError

0.43

 Geſch

0.43

Activations Density 0.013%

references to data structure manipulations and programming-related terms

No Comments

No Known Activations