INDEX

Explanations

information related to secure environments and notifications regarding policy changes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.27.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

 oddly

-0.41

 somewhat

-0.40

 strangely

-0.39

 faintly

-0.39

(?)

-0.39

 infamous

-0.38

 suddenly

-0.37

TryParse

-0.36

(?)

-0.35

 bizarre

-0.35

POSITIVE LOGITS

 myſelf

0.85

ſelves

0.80

 purpoſe

0.79

 Houſe

0.78

 ſch

0.76

BibitemShut

0.76

 auffi

0.72

 Reſ

0.71

 houſe

0.71

ſelf

0.71

Activations Density 2.194%

information related to secure environments and notifications regarding policy changes

No Comments

No Known Activations