INDEX

Explanations

negative descriptors and criticisms related to individuals and situations

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

negative descriptors related to treatment and behavior

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_mlp_out

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

 Worse

-1.07

Worse

-1.07

 peor

-1.03

 worse

-1.00

 peores

-0.93

 poorer

-0.92

😥

-0.91

😒

-0.91

 poorly

-0.90

 poor

-0.90

POSITIVE LOGITS

 <<<<<<<<<<<<<<

0.76

 exceptional

0.67

脚注の使い方

0.65

➟

0.62

 uniquely

0.62

setVerticalGroup

0.61

 estup

0.60

Personendaten

0.60

 FontWeight

0.59

 ویکی‌پدیا

0.59

Activations Density 2.920%

negative descriptors and criticisms related to individuals and situations

negative descriptors related to treatment and behavior

No Comments

No Known Activations