INDEX

Explanations

terms related to programming functions and legal documentation

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

patient consent and code contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 always

-0.42

 vedno

-0.40

 altijd

-0.39

 zawsze

-0.39

Reprodução

-0.37

 мѣ

-0.36

always

-0.35

setParent

-0.35

 toujours

-0.33

 kurulan

-0.33

POSITIVE LOGITS

 propOrder

0.74

 ujednoznacz

0.69

 للمعارف

0.68

RenderAtEndOf

0.59

әрмәләр

0.59

ReusableCell

0.57

Hochspringen

0.56

ьаж

0.56

haikusbot

0.54

parsedMessage

0.53

Activations Density 0.000%

terms related to programming functions and legal documentation

patient consent and code contexts

No Comments

No Known Activations