INDEX

Explanations

references to programming and class structure in code

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

vestream

-0.48

 timedelta

-0.47

EDEFAULT

-0.46

évé

-0.45

 оригіналу

-0.44

 PyQt

-0.43

 Biôgrafia

-0.42

autogui

-0.42

ústria

-0.41

 fevere

-0.41

POSITIVE LOGITS

quelize

0.48

modelBuilder

0.48

InitVars

0.41

protoc

0.41

EndInit

0.40

CreateModel

0.39

دانشنامهٔ

0.39

OpenHelper

0.39

CreateTable

0.38

MemoryWarning

0.38

Activations Density 0.003%

references to programming and class structure in code

No Comments

No Known Activations