INDEX

Explanations

punctuation and structure in text

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

medical terms after 'and'

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

TagMode

-0.82

ViewImports

-0.75

WriteTagHelper

-0.74

BufferException

-0.74

पया

-0.70

IsMutable

-0.69

 дописавши

-0.67

цездатний

-0.67

السكان

-0.66

 يتيمه

-0.64

POSITIVE LOGITS

 also

0.43

clusal

0.37

 также

0.36

 ayrıca

0.36

Also

0.36

 lisäksi

0.36

also

0.35

too

0.35

 myös

0.35

 Also

0.35

Activations Density 0.003%

punctuation and structure in text

medical terms after 'and'

No Comments

No Known Activations