INDEX

Explanations

programming constructs related to loops and conditional statements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

code, mathematical notation

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WR

-0.46

 Predic

-0.40

JNIEnv

-0.40

 Accept

-0.39

⊃

-0.38

Clik

-0.38

ṟ

-0.38

 rede

-0.38

 camicia

-0.38

Sip

-0.37

POSITIVE LOGITS

SourceChecksum

0.66

 ExecuteAsync

0.51

GenerationType

0.50

 HasFactory

0.49

 itſelf

0.48

 establecimiento

0.46

yntaxException

0.45

 GenerationType

0.45

Soorten

0.43

 raiſ

0.43

Activations Density 0.032%

programming constructs related to loops and conditional statements

code, mathematical notation

No Comments

No Known Activations