INDEX

Explanations

patterns of decision-making and actions taken by individuals in various scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/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.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

Противо

-0.50

よろしくお願いします

-0.47

tilus

-0.47

 gefördert

-0.45

ंत्र

-0.45

 TextAppearance

-0.45

MessageTagHelper

-0.44

 alternation

-0.44

익

-0.44

rony

-0.43

POSITIVE LOGITS

 decided

2.73

 decide

2.54

 decides

2.41

decided

2.26

 Decided

2.21

 decision

2.13

 décide

2.08

 Decide

2.07

 deciding

2.05

decide

2.04

Activations Density 0.579%

patterns of decision-making and actions taken by individuals in various scenarios

No Comments

No Known Activations