INDEX

Explanations

words that have alternative valences, such as close/far, or up/down.

oai_token-act-pair · gemini-2.0-flash

Proximity/closeness

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-0.72

 picioare

-0.70

 hvid

-0.57

 estadía

-0.56

 suivantes

-0.56

 chiuso

-0.55

 spéciaux

-0.54

 maș

-0.54

 condamné

-0.54

 forcé

-0.54

POSITIVE LOGITS

colour

0.97

initialise

0.94

Colours

0.93

colours

0.91

Whilst

0.91

 apologise

0.89

 Whilst

0.88

 colours

0.86

 colour

0.85

COLOUR

0.84

Activations Density 1.879%

words that have alternative valences, such as close/far, or up/down.

Proximity/closeness

No Comments

No Known Activations