INDEX

Explanations

programming-related terms and operations, especially those involving data structures and algorithms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

<eos>

-0.45

 sheer

-0.43

片

-0.41

 path

-0.40

BeginInit

-0.40

mada

-0.40

льность

-0.40

__":

-0.40

gerichts

-0.39

oneofs

-0.39

POSITIVE LOGITS

UnusedPrivate

0.77

TagMode

0.75

 useState

0.73

 bezeichneter

0.72

queryInterface

0.71

 الاطلاع

0.69

 snippetHide

0.67

useState

0.66

InitVars

0.65

bkz

0.64

Activations Density 0.231%

programming-related terms and operations, especially those involving data structures and algorithms

No Comments

No Known Activations