INDEX

Explanations

terms related to previous occurrences or historical context

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

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

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Zjednoc

-0.65

 odkazy

-0.55

 ralenti

-0.55

 crudo

-0.55

 envies

-0.55

IVATE

-0.54

ĩa

-0.54

 vivos

-0.54

 articula

-0.54

 gravi

-0.54

POSITIVE LOGITS

SourceChecksum

0.68

 AssemblyCulture

0.66

ніципалі

0.65

})*/

0.63

 CYTHON

0.62

']").

0.62

 seen

0.62

꿔

0.59

PhysRev

0.59

 universally

0.58

Activations Density 0.083%

terms related to previous occurrences or historical context

No Comments

No Known Activations