INDEX

Explanations

phrases related to political investigations and accountability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

脚注の使い方

-0.53

 ekst

-0.45

Annotations

-0.45

^^^^^^^^

-0.45

הערות

-0.42

align

-0.41

ător

-0.41

ịp

-0.40

diez

-0.40

 kır

-0.39

POSITIVE LOGITS

 investigation

2.45

 investigations

2.17

investigation

2.11

 Investigation

2.06

 investigating

2.04

Investigation

1.99

 investigate

1.98

investig

1.93

 INVESTIGATION

1.89

 investigates

1.85

Activations Density 0.372%

phrases related to political investigations and accountability

No Comments

No Known Activations