INDEX

Explanations

discussions that involve legal terminology or jurisprudence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

was

-0.42

Reference

-0.33

 sempat

-0.32

reference

-0.31

 było

-0.31

 fikir

-0.30

spot

-0.30

substance

-0.30

 spot

-0.30

↵↵

-0.30

POSITIVE LOGITS

 الحره

0.74

 فريبيس

0.65

DockStyle

0.65

.*")]

0.62

addContainerGap

0.62

رشف

0.59

 مشين

0.59

#+#

0.59

AnchorTagHelper

0.59

хьтан

0.59

Activations Density 1.510%

discussions that involve legal terminology or jurisprudence

No Comments

No Known Activations