INDEX

Explanations

code-related keywords and class definitions

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

ftagPool

-0.59

RegressionTest

-0.56

MLLoader

-0.53

новништво

-0.50

":[{

-0.49

AsUp

-0.48

 المعيارى

-0.48

 Biôgrafia

-0.48

ThroughAttribute

-0.47

stylers

-0.46

POSITIVE LOGITS

getSimpleName

0.60

iibo

0.44

tableFuture

0.42

₿

0.42

 stoichio

0.42

 dumpster

0.42

ostasis

0.41

 clazz

0.41

SpringBootTest

0.41

 viewDidLoad

0.41

Activations Density 0.000%

code-related keywords and class definitions

No Comments

No Known Activations