INDEX

Explanations

expressions of self-identity and personal characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 desconhe

-0.61

 neceffary

-0.58

 Jefus

-0.57

ResponseWriter

-0.57

 preſent

-0.57

 subordinated

-0.56

 unknowns

-0.55

 myſelf

-0.55

 hostilities

-0.52

 Mongols

-0.52

POSITIVE LOGITS

 notoriously

0.92

 notorious

0.89

 brukar

0.73

 prone

0.73

 famously

0.73

 known

0.72

 always

0.71

 usually

0.71

 often

0.70

 obsessive

0.69

Activations Density 0.232%

expressions of self-identity and personal characteristics

No Comments

No Known Activations