INDEX

Explanations

programming syntax and structure related to variable assignments and function definitions

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

hom warp fetch pci user

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 bowiem

-0.48

StructEnd

-0.45

EDEFAULT

-0.45

 przecież

-0.44

 vrijwilli

-0.44

 indeed

-0.43

indeed

-0.42

 avoient

-0.41

randomUUID

-0.41

 kenyataan

-0.41

POSITIVE LOGITS

Билгалдахарш

0.68

WriteLiteral

0.62

setVerticalGroup

0.61

 Bulgarian

0.54

Bal

0.52

 stitch

0.52

 Ajax

0.52

<%@

0.52

})));

0.50

"}>

0.50

Activations Density 0.002%

programming syntax and structure related to variable assignments and function definitions

hom warp fetch pci user

No Comments

No Known Activations