INDEX

Explanations

references to URLs or digital object identifiers (DOIs)

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

twimg

-0.94

 ***!

-0.88

 препратки

-0.85

 GenerationType

-0.82

 ModelExpression

-0.82

DOCTYPE

-0.79

 springfox

-0.78

AndEndTag

-0.76

 getItemId

-0.73

protoc

-0.73

POSITIVE LOGITS

 ſch

0.71

 Romains

0.65

 dafs

0.63

 varmt

0.61

 ſur

0.59

 Inscrivez

0.58

 Reſ

0.57

Eſ

0.57

 auffi

0.57

 tranſ

0.57

Activations Density 0.001%

references to URLs or digital object identifiers (DOIs)

No Comments

No Known Activations