INDEX

Explanations

references to legal hearings and proceedings

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

tiérrez

-0.53

StateToProps

-0.50

colorado

-0.49

itsubishi

-0.48

 Meksika

-0.48

 cucchiai

-0.47

productName

-0.46

 inspirasi

-0.46

tiglia

-0.46

﹍﹍﹍

-0.45

POSITIVE LOGITS

 hearings

0.82

 Hearings

0.82

 Hearing

0.69

 hearing

0.64

Hearing

0.59

HEARING

0.57

hearing

0.57

 Meeting

0.55

hearsal

0.54

quium

0.52

Activations Density 0.018%

references to legal hearings and proceedings

No Comments

No Known Activations