INDEX

Explanations

terms related to standards and common practices in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 larmes

-0.46

 jambe

-0.46

 siihen

-0.43

 tertawa

-0.42

nahme

-0.42

domada

-0.42

 basah

-0.41

fertigt

-0.40

 hitam

-0.40

phemy

-0.40

POSITIVE LOGITS

 modern

0.61

 contemporary

0.52

 ExecuteAsync

0.49

LookAnd

0.48

modern

0.46

 informée

0.46

 cultures

0.45

 civilizations

0.44

ByVersion

0.44

 modernen

0.44

Activations Density 0.602%

terms related to standards and common practices in various contexts

No Comments

No Known Activations