INDEX

Explanations

specific nouns and terms related to processes and objects 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

readFileSync

-0.38

 ích

-0.33

 intersection

-0.33

 Thorne

-0.32

 omit

-0.32

ioutil

-0.32

 Grüße

-0.31

วน

-0.31

 lombok

-0.30

Printf

-0.30

POSITIVE LOGITS

writeFieldEnd

0.63

 AssemblyVersion

0.63

RenderAtEndOf

0.60

tvguidetime

0.58

 Photocase

0.57

 дописавши

0.57

WebElementEntity

0.56

httphttps

0.54

ValueStyle

0.53

Personendaten

0.53

Activations Density 0.877%

specific nouns and terms related to processes and objects in various contexts

No Comments

No Known Activations