INDEX

Explanations

interrogative and conjunction words in questions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 grun

-0.72

 Chriſt

-0.70

 Estou

-0.70

awtextra

-0.68

homem

-0.68

 purpoſe

-0.67

 eſt

-0.67

empereur

-0.66

hip

-0.66

 westfalen

-0.66

POSITIVE LOGITS

Que

0.93

Que

0.92

__":

0.85

que

0.85

QUE

0.80

__':

0.80

'];

0.77

__':

0.76

AutoField

0.75

″]

0.74

Activations Density 0.020%

interrogative and conjunction words in questions

No Comments

No Known Activations