INDEX

Explanations

the context references in programming-related documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 sánh

-0.62

 CreateTagHelper

-0.57

rusted

-0.56

bah

-0.55

ca

-0.55

ypso

-0.54

arache

-0.52

belline

-0.52

RFI

-0.52

xl

-0.52

POSITIVE LOGITS

 Orientation

0.75

Orientation

0.72

 Majefty

0.69

 Houſe

0.68

orientation

0.68

 Oriented

0.67

 occaf

0.66

 neck

0.64

 reafon

0.64

 orientation

0.64

Activations Density 0.052%

the context references in programming-related documentation

No Comments

No Known Activations