INDEX

Explanations

the repeated use of the word "so" in various contexts, indicating emphasis or intensity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_13/width_16k/average_l0_132

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.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

 elástica

-0.53

Ethylbenzene

-0.52

ruptcy

-0.51

 trasera

-0.48

rugada

-0.47

utzung

-0.47

 eficaz

-0.46

QUIS

-0.45

いない

-0.45

 matelas

-0.45

POSITIVE LOGITS

so

0.98

So

0.93

SO

0.93

So

0.90

so

0.79

Sooo

0.79

SO

0.75

Soo

0.67

ſo

0.66

Soo

0.66

Activations Density 0.166%

the repeated use of the word "so" in various contexts, indicating emphasis or intensity

No Comments

No Known Activations