INDEX

Explanations

terms related to legal judgments and court proceedings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 surla

-0.84

ArrowToggle

-0.76

iastes

-0.72

MigrationBuilder

-0.68

WireFormatLite

-0.68

 Elisha

-0.66

PullParser

-0.64

 Ferner

-0.62

:\/\/

-0.61

igraph

-0.61

POSITIVE LOGITS

 guilty

1.90

guilty

1.70

 Guilty

1.59

 conviction

1.33

Guilty

1.31

 convicted

1.24

 convictions

1.22

 convic

1.09

 guilt

1.02

 conscious

0.99

Activations Density 0.175%

terms related to legal judgments and court proceedings

No Comments

No Known Activations