INDEX

Explanations

code-related concepts and structures, particularly those associated with object properties and methods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 ſeveral

-0.86

 tranſ

-0.84

 виправивши

-0.82

 auffi

-0.82

 ſtate

-0.81

Autoritní

-0.81

 ſever

-0.79

 Jefus

-0.79

 uſed

-0.78

 läßt

-0.78

POSITIVE LOGITS

checkIf

0.60

 entirety

0.56

Alright

0.54

numOf

0.51

nanoTime

0.51

fName

0.51

 rightfully

0.50

esque

0.50

除此之外

0.50

 aforementioned

0.49

Activations Density 0.189%

code-related concepts and structures, particularly those associated with object properties and methods

No Comments

No Known Activations