INDEX

Explanations

references to government actions and criticisms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_19/width_16k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.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

DebuggerNonUser

-0.66

 ridiculous

-0.62

 precious

-0.62

 précie

-0.61

 Chwiliwch

-0.59

 cowardly

-0.58

 pitiful

-0.57

 pointless

-0.56

Lähteet

-0.56

 valuable

-0.55

POSITIVE LOGITS

 apparently

0.79

apparently

0.75

 Apparently

0.66

Apparently

0.64

 allegedly

0.62

 aparentemente

0.62

 apparente

0.59

WithIOException

0.59

 reportedly

0.57

 seemingly

0.56

Activations Density 0.679%

references to government actions and criticisms

No Comments

No Known Activations