INDEX

Explanations

pronouns and references to entities in a discourse context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 complex

-0.53

 isolato

-0.53

 complicated

-0.50

 Consiglio

-0.48

 individuale

-0.48

 Complex

-0.47

 ilişk

-0.46

 complexes

-0.46

 regola

-0.46

 riserv

-0.45

POSITIVE LOGITS

Personensuche

1.04

 utafitiHapana

0.91

脚注の使い方

0.90

 تانيه

0.81

 للمعارف

0.81



0.80

المناصب

0.78

__(/*!

0.76

istoitu

0.74

IndentedString

0.70

Activations Density 0.605%

pronouns and references to entities in a discourse context

No Comments

No Known Activations