INDEX

Explanations

the author's name or credit in a document

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

user turn with code

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

 memor

-0.43

 Memor

-0.42

lesias

-0.42

ngrx

-0.41

OwnProperty

-0.41

DNEY

-0.40

 still

-0.39

sca

-0.39

nestjs

-0.37

 terrain

-0.37

POSITIVE LOGITS

 AssemblyCompany

0.82

 AssemblyVersion

0.70

 AssemblyTitle

0.68

ViewFeatures

0.68

Personendaten

0.61

hyrchwyd

0.57

 ComVisible

0.56

 AssemblyProduct

0.55

WithMany

0.52

\{\\

0.51

Activations Density 0.000%

the author's name or credit in a document

user turn with code

No Comments

No Known Activations