INDEX

Explanations

phrases expressing assumptions or presumption

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nemlig

-0.64

 autorytatywna

-0.64

 Majefty

-0.63

 betweenstory

-0.62

Spécification

-0.60

ftance

-0.60

spesies

-0.60

empêcher

-0.59

WebElementEntity

-0.58

 BorderSide

-0.58

POSITIVE LOGITS

 probably

0.65

 assume

0.64

 presumably

0.60

umably

0.57

Probably

0.57

 Probably

0.56

 Presumably

0.54

probably

0.53

presumably

0.52

assume

0.50

Activations Density 0.036%

phrases expressing assumptions or presumption

No Comments

No Known Activations