INDEX

Explanations

phrases indicating skepticism or doubt about claims and experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

ſelf

-0.50

 Tinggi

-0.48

 Monfieur

-0.44

 Pembangunan

-0.44

ždy

-0.43

 Kecil

-0.43

 насељу

-0.42

ěstí

-0.42

 Kerja

-0.42

 Lagi

-0.42

POSITIVE LOGITS

 even

0.76

 siquiera

0.73

even

0.68

Even

0.62

 навіть

0.61

Tembelea

0.60

 даже

0.60

 Even

0.57

 barely

0.57

 EVEN

0.56

Activations Density 0.139%

phrases indicating skepticism or doubt about claims and experiences

No Comments

No Known Activations