INDEX

Explanations

technical terms related to computational tasks and network programming

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

NewErrorWithError

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

Архівовано

-0.47

 Italijanski

-0.40

 BufferedReader

-0.37

)][

-0.36

 Jesucristo

-0.36

 hence

-0.36

πή

-0.35

Glej

-0.34

unikasi

-0.34

 afterwards

-0.34

POSITIVE LOGITS

 surla

0.66

WriteTagHelper

0.56

:✨

0.55

featureID

0.53

 invokingState

0.52

ècie

0.52

 TestBed

0.50

localctx

0.49

DoubleQuotes

0.47

ביוגרפיה

0.47

Activations Density 0.002%

technical terms related to computational tasks and network programming

NewErrorWithError

No Comments

No Known Activations