INDEX

Explanations

terms related to citations and data retrieval

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_32k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.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

ministrazione

-0.63

LEADING

-0.60

baguna

-0.59

 createTime

-0.54

kuuta

-0.54

('../../

-0.53

 Cummins

-0.52

Precios

-0.52

tiempo

-0.51

fieldname

-0.51

POSITIVE LOGITS

 Consultado

1.54

 abgerufen

1.37

 Процитовано

1.20

 оригіналу

1.19

 Abgerufen

1.17

 consultato

1.13

consulté

1.12

 Retrieved

1.10

 Diakses

1.08

 Архивировано

1.05

Activations Density 0.005%

terms related to citations and data retrieval

No Comments

No Known Activations