INDEX

Explanations

references to entities that are self-referential or circular in nature

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 hunne

-0.65

 zijne

-0.58

ourable

-0.58

 whoſe

-0.57

 Nuestros

-0.57

Kenzie

-0.55

 dolayı

-0.55

rdı

-0.54

 Reds

-0.54

 enrique

-0.54

POSITIVE LOGITS

 itself

1.67

itself

1.63

 Itself

1.45

themselves

1.19

 himself

1.14

 themselves

1.14

 Himself

1.10

 herself

1.09

 sendiri

1.06

herself

1.04

Activations Density 0.107%

references to entities that are self-referential or circular in nature

No Comments

No Known Activations