INDEX

Explanations

titles and ranks of characters or individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.35.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

tagHelperRunner

-0.63

Tembelea

-0.47

iedenen

-0.45

 unconditionally

-0.43

ävän

-0.42

 SEDS

-0.42

 kaynağından

-0.41

 presumably

-0.41

쿵

-0.41

 podjela

-0.41

POSITIVE LOGITS

########.

0.71

enderror

0.55

 AssemblyTitle

0.53

 Emeritus

0.45

们

0.45

們

0.43

们

0.42

bibitem

0.42

 John

0.42

jspb

0.42

Activations Density 0.282%

titles and ranks of characters or individuals

No Comments

No Known Activations