INDEX

Explanations

references to programming variables and their characteristics

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

technical and scientific contexts

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

Cyfarwyddwr

-0.63

 humanidad

-0.48

 leña

-0.48

 jagung

-0.48

Diweddarwch

-0.47

Creator

-0.47

 oreja

-0.47

 antigüedad

-0.47

 '\\;'

-0.47

 herido

-0.45

POSITIVE LOGITS

Билгалдахарш

0.53

帖最后由

0.49

TagMode

0.45

utilisons

0.41

transQ

0.39

pushFollow

0.39

mphs

0.37

ंदीखरीदारी

0.36

WithIOException

0.35

weightedMode

0.35

Activations Density 0.012%

references to programming variables and their characteristics

technical and scientific contexts

No Comments

No Known Activations