INDEX

Explanations

phrases related to self-interest and exploitation in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

zep

-0.50

 juridiques

-0.47

 next

-0.45

UnifiedTopology

-0.44

rá

-0.44

jScrollPane

-0.42

tillon

-0.42

SCI

-0.41

 nueces

-0.41

چ

-0.41

POSITIVE LOGITS

 متعلقه

0.82

PreferredItem

0.81

Portale

0.78

 autorytatywna

0.76

 motives

0.74

 <=",

0.72

parsedMessage

0.72

RegressionTest

0.71

 motive

0.70

 للاسماء

0.69

Activations Density 0.401%

phrases related to self-interest and exploitation in various contexts

No Comments

No Known Activations