INDEX

Explanations

references to image processing and data structures in programming contexts

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

 otomatig

-0.47

 argint

-0.44

 poważ

-0.44

 šķ

-0.43

 düğ

-0.42

 hueso

-0.41

 odbior

-0.41

 detener

-0.40

 agencia

-0.39

 franquicia

-0.39

POSITIVE LOGITS

Xna

0.60

///</

0.56

pushFollow

0.45

ⓧ

0.44

ContentAlignment

0.44

enterOuterAlt

0.43

 genomen

0.41

AutoScaleMode

0.41

 kasarigan

0.41

 sank

0.41

Activations Density 0.058%

references to image processing and data structures in programming contexts

No Comments

No Known Activations