INDEX

Explanations

terms and references related to academic or technical formats and structures

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

here, seen, FASTA format

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

出版年

-0.60

tonsoft

-0.58

MemoryWarning

-0.57

 protoimpl

-0.56

Obrigada

-0.52

Tikang

-0.51

كويكب

-0.50

 obispo

-0.50

MLLoader

-0.48

}{*}{

-0.48

POSITIVE LOGITS

 wireType

0.38

 therefore

0.36

RuleContext

0.36

 sometimes

0.35

WireFormatLite

0.35

 here

0.34

loc

0.33

 cantos

0.33

fel

0.31

LLocation

0.31

Activations Density 0.265%

terms and references related to academic or technical formats and structures

here, seen, FASTA format

No Comments

No Known Activations