INDEX

Explanations

phrases indicating confirmation or agreement in discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 8-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_8/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.8.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

 expense

-0.54

expense

-0.51

あなたが

-0.46

 complex

-0.45

inander

-0.45

しまいます

-0.45

 المعيارى

-0.45

 hjelp

-0.44

うちに

-0.44

 entrada

-0.44

POSITIVE LOGITS

TintMode

0.80

ValueStyle

0.78

StoryboardSegue

0.77

parsedMessage

0.75

 referenties

0.69

 ویکی‌آمباردا

0.67

LEncoder

0.66

 preached

0.64

ufo

0.62

CPtr

0.62

Activations Density 0.241%

phrases indicating confirmation or agreement in discussions

No Comments

No Known Activations