INDEX

Explanations

evaluative language that indicates effectiveness or success

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

Efq

-0.76

 Monfieur

-0.74

Eſ

-0.71

 Jefus

-0.71

 perfons

-0.69

 perſon

-0.68

 ſeveral

-0.67

matchCondition

-0.67

setVerticalGroup

-0.66

цездатний

-0.66

POSITIVE LOGITS

SizeMode

0.54

EFFECT

0.51

rativa

0.47

 Consultado

0.47

 gratify

0.47

Петербург

0.46

roit

0.45

GIVEREF

0.45

asteroide

0.44

 effek

0.44

Activations Density 0.278%

evaluative language that indicates effectiveness or success

No Comments

No Known Activations