INDEX

Explanations

references to legal proceedings and allegations

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 hata

-0.27

-0.26

Et

-0.26

Hat

-0.25

 American

-0.25

 iprot

-0.25

offset

-0.24

str

-0.24

sn

-0.24

 русских

-0.24

POSITIVE LOGITS

contentLoaded

0.83

parsedMessage

0.67

ロウィン

0.61

tonode

0.61

تقاوى

0.60

 prosecutors

0.60

𐑥

0.59

࿊

0.59

RenderAtEndOf

0.59

UniformLocation

0.59

Activations Density 0.379%

references to legal proceedings and allegations

No Comments

No Known Activations