INDEX

Explanations

references to physical structures or locations, particularly buildings and apartments

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

webElementXpaths

-0.45

 useCallback

-0.42

 становника

-0.40

zzá

-0.40

IndexPath

-0.39

windowFixed

-0.39

 NSError

-0.39

IntoConstraints

-0.38

ceğine

-0.38

tableFuture

-0.37

POSITIVE LOGITS

ExtendWith

0.50

団

0.50

 building

0.50

RTDA

0.49

 Building

0.47

tower

0.46

 AssemblyVersion

0.44

Building

0.44

 tower

0.44

 Buildings

0.44

Activations Density 0.009%

references to physical structures or locations, particularly buildings and apartments

No Comments

No Known Activations