INDEX

Explanations

references to solutions and improvements

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

the word "solution"

np_acts-logits-general · gemini-2.0-flash

discussing a solution

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

 Orte

-0.51

tuan

-0.49

wain

-0.48

 Zeichen

-0.48

madre

-0.48

disi

-0.47

onucleotide

-0.47

 AssemblyCompany

-0.47

 Warning

-0.47

鰭

-0.47

POSITIVE LOGITS

 provider

0.75

 Provider

0.69

 providers

0.64

writeFieldEnd

0.64

provider

0.63

 Providers

0.61

 TestBed

0.58

 löyty

0.58

Provider

0.57

ingen

0.54

Activations Density 0.182%

references to solutions and improvements

the word "solution"

discussing a solution

No Comments

No Known Activations