INDEX

Explanations

mentions of judicial errors or issues in legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 الرياضيه

-1.20

 يتيمه

-1.14

 Administrativna

-1.13

#+#

-1.10

 typelib

-1.09

RenderAtEndOf

-1.07

tagHelperRunner

-1.06

 otomatig

-1.04

 AssemblyCulture

-1.03

出版年

-1.03

POSITIVE LOGITS

Finally

0.65

 finally

0.65

 Finally

0.63

↵↵

0.58

<h2>

0.57

↵

0.56

<em>

0.53

 lastly

0.52

Lastly

0.52

finally

0.51

Activations Density 0.302%

mentions of judicial errors or issues in legal contexts

No Comments

No Known Activations