INDEX

Explanations

elements related to querying and data retrieval in programming or database contexts

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

numbers and math notation

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

PMailer

-0.66

 disambiguazione

-0.65

OGND

-0.62

 BoxFit

-0.60

LabelTagHelper

-0.59

jsonwebtoken

-0.57

rawDesc

-0.57

gameserver

-0.56

RenderAtEndOf

-0.55

AutoScaleMode

-0.55

POSITIVE LOGITS

føl

0.47

 corrientes

0.39

 desnuda

0.37

 almohada

0.36

 aislada

0.35

 valiente

0.35

 controlada

0.35

 práctico

0.34

 evolución

0.34

 colgante

0.34

Activations Density 0.090%

elements related to querying and data retrieval in programming or database contexts

numbers and math notation

No Comments

No Known Activations