INDEX

Explanations

phrases related to actions and procedures within technical documents

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

 Monfieur

-0.90

 myſelf

-0.88

 himſelf

-0.85

 itſelf

-0.84

 AssemblyVersion

-0.79

 purpoſe

-0.78

 ſen

-0.77

}();

-0.76

InjectAttribute

-0.76

ſelf

-0.75

POSITIVE LOGITS

 usually

0.62

 often

0.61

 always

0.56

 sometimes

0.54

usually

0.54

 Usually

0.53

 zawsze

0.52

Usually

0.51

 vždy

0.51

often

0.50

Activations Density 0.415%

phrases related to actions and procedures within technical documents

No Comments

No Known Activations

phrases related to actions and procedures within technical documents

No Comments

No Known Activations