INDEX

Explanations

conjunctions or phrases that suggest relationships between concepts or lists

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.17.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

Datuak

-0.71

KommentareTeilen

-0.67

 الاطلاع

-0.66

Eloquent

-0.65

SharedDtor

-0.65

 AssemblyVersion

-0.64

CloseOperation

-0.63

 StatefulWidget

-0.61

assas

-0.60

jonal

-0.59

POSITIVE LOGITS

 lesquels

0.65

 that

0.65

had

0.63

are

0.61

 magari

0.61

ViewFeatures

0.55

 whether

0.54

who

0.53

basicConfig

0.53

 whose

0.51

Activations Density 0.378%

conjunctions or phrases that suggest relationships between concepts or lists

No Comments

No Known Activations