INDEX

Explanations

terms related to recognition and awareness

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

Different forms of "recognize" and "ignore"

np_acts-logits-general · gemini-2.0-flash

recognize, recognition, acknowledging, ignored

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

/\.

-0.59

########.

-0.48

Ara

-0.40

 CanadaChoose

-0.40

Islands

-0.39

 myſelf

-0.39

saraba

-0.38

 ISLANDS

-0.37

Ara

-0.36

leste

-0.35

POSITIVE LOGITS

 recognize

0.93

 Recognize

0.93

 recognise

0.93

 recognises

0.88

 Recognition

0.87

 recognizes

0.86

Recognize

0.84

 recognition

0.84

recognize

0.84

 Recognized

0.83

Activations Density 0.137%

terms related to recognition and awareness

Different forms of "recognize" and "ignore"

recognize, recognition, acknowledging, ignored

No Comments

No Known Activations