INDEX

Explanations

terms and concepts related to algorithms and programming structures

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 للاسماء

-0.68

 الحره

-0.63

 utafitiHapana

-0.57

 OMITBAD

-0.56

lankton

-0.56

Personendaten

-0.55

uruhan

-0.52

\{\\

-0.51

rétaire

-0.51

cadilly

-0.50

POSITIVE LOGITS

 techniques

0.35

 technique

0.34

Skocz

0.33

mstyle

0.32

Hol

0.32

 metode

0.31

teknik

0.31

 structure

0.31

 struttura

0.31

 dày

0.31

Activations Density 0.264%

terms and concepts related to algorithms and programming structures

No Comments

No Known Activations