INDEX

Explanations

references to emotional manipulation and psychological conflict

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 Majefty

-0.83

 Cæsar

-0.81

 pleaſure

-0.71

 themſelves

-0.69

 itſelf

-0.69

 Shakspeare

-0.68

 Galile

-0.68

 ་་

-0.68

 electrico

-0.66

 Zoro

-0.66

POSITIVE LOGITS

 everyone

0.85

 everybody

0.80

everyone

0.75

everybody

0.74

Mr

0.73

TintMode

0.73

 others

0.70

Dr

0.68

 someone

0.68

MigrationBuilder

0.68

Activations Density 0.450%

references to emotional manipulation and psychological conflict

No Comments

No Known Activations