INDEX

Explanations

alphanumeric codes or identifiers, particularly within a technical or data-related context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

)";

-0.70



-0.67

 ")";

-0.66

RRect

-0.60

etur

-0.56

DrawerToggle

-0.56

InitVars

-0.55

issimo

-0.54

 <>",

-0.54

Efq

-0.54

POSITIVE LOGITS

出版年

0.60

 تعدى

0.56

šní

0.52

PerformLayout

0.52

 nicio

0.50

 ști

0.49

MILLIS

0.47

 possano

0.46

 arqu

0.46

 aleación

0.46

Activations Density 0.082%

alphanumeric codes or identifiers, particularly within a technical or data-related context

No Comments

No Known Activations