INDEX

Explanations

phrases and terminology related to legal standards and procedural norms

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RenderAtEndOf

-0.55

بوابة

-0.52

HtmlAttribute

-0.52

idom

-0.45

Tikang

-0.45

gms

-0.44

providedIn

-0.41

 verf

-0.41

TypedDataSet

-0.41

Externé

-0.41

POSITIVE LOGITS

 Umum

0.60

 vaksinasi

0.52

 numerus

0.51

Usually

0.49

 umumnya

0.48

 Informasi

0.47

 normalt

0.47

ทั่ว

0.47

 auffi

0.47

 Usually

0.47

Activations Density 0.057%

phrases and terminology related to legal standards and procedural norms

No Comments

No Known Activations