INDEX

Explanations

sections of structured data or formatted information, likely including numerical or coded elements

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

Formatting boundaries or specific elements within structured data representations like arrays, tables, or key-value dictionaries, often appearing at line transitions or as complete data entries.

eleuther_acts_top20 · claude-3-7-sonnet-20250219 Triggered by @lu-christina

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 noDo

-0.76

 transfieras

-0.75

 ویکی‌پدی

-0.71

 queſta

-0.71

PyExc

-0.71

 tartalomajánló

-0.70

SharedDtor

-0.70

fromnode

-0.68

ьажоргаш

-0.68

 lenker

-0.66

POSITIVE LOGITS

↵↵

0.37

0.35

 record

0.34

0.33

and

0.33

↵

0.33

to

0.32

 parç

0.31

Activations Density 0.378%

sections of structured data or formatted information, likely including numerical or coded elements

Formatting boundaries or specific elements within structured data representations like arrays, tables, or key-value dictionaries, often appearing at line transitions or as complete data entries.

No Comments

No Known Activations