INDEX

Explanations

mentions of interviews and documents related to investigative reporting

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

former Apple engineers

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

 Whilst

-0.44

 whilst

-0.42

Whilst

-0.42

Tikang

-0.39

 rumours

-0.38

 films

-0.38

 Club

-0.37

writerow

-0.37

 Films

-0.36

 galleries

-0.36

POSITIVE LOGITS

 HasFactory

0.59

SpringRunner

0.53

TemporalType

0.53

 TestBed

0.52

 poffible

0.51

ſelf

0.50

rapnel

0.50

konomi

0.49

 TextInputType

0.48

cade

0.47

Activations Density 0.073%

mentions of interviews and documents related to investigative reporting

former Apple engineers

No Comments

No Known Activations