INDEX

Explanations

descriptive adjectives and phrases that convey positivity and complexity

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

rungsseite

-0.68

VersionUID

-0.49

 feroit

-0.49

 financeira

-0.49

pulseira

-0.49

CppMethod

-0.48

 themselves

-0.48

caneca

-0.47

 koncep

-0.47

 harita

-0.47

POSITIVE LOGITS

 Савезне

0.46

 poffible

0.37

ConstraintMaker

0.37

 noten

0.36

boer

0.36

ISI

0.36

 indes

0.35

''

0.35

Portail

0.34

 ModelRenderer

0.34

Activations Density 0.224%

descriptive adjectives and phrases that convey positivity and complexity

No Comments

No Known Activations