INDEX

Explanations

coding-related terms and commands

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

Personendaten

-0.53

viola

-0.41

pexpr

-0.40

Coup

-0.39

DebuggerNonUser

-0.38

Her

-0.37

 automatic

-0.36

 MediaQuery

-0.36

gatsby

-0.36

plu

-0.35

POSITIVE LOGITS

StackNavigator

0.62

rungsseite

0.56

OptionsMenu

0.54

urlpatterns

0.52

PARATUS

0.49

modelBuilder

0.47

pecabe

0.46

 NgModule

0.45

antMatchers

0.44

IntoConstraints

0.44

Activations Density 0.047%

coding-related terms and commands

No Comments

No Known Activations