INDEX

Explanations

mathematical notation and formal expressions related to functions and sequences

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

mathematical sums and sequences

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

 kasarigan

-0.56

PMailer

-0.53

 intStringLen

-0.52

URLException

-0.50

InputBorder

-0.44

matchCondition

-0.44

 työn

-0.44

 CreateTagHelper

-0.43

Pozri

-0.43

andafter

-0.43

POSITIVE LOGITS

 informée

0.40

expandindo

0.38

XmlEnum

0.35

saco

0.35

 undocumented

0.34

 Cresc

0.34

 geni

0.33

Gegen

0.33

 muda

0.32

 khuẩn

0.31

Activations Density 0.121%

mathematical notation and formal expressions related to functions and sequences

mathematical sums and sequences

No Comments

No Known Activations