INDEX

Explanations

phrases that denote rankings or comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 taip

-0.35

nymi

-0.32

tamp

-0.31

borgen

-0.31

 Plus

-0.28

 until

-0.28

forward

-0.28

ưởng

-0.28

dataGridView

-0.28

 absolutely

-0.28

POSITIVE LOGITS

 nakalista

0.81

 AssemblyCulture

0.71

enderror

0.69

 Normdatei

0.68

 ***!

0.66

出版年

0.66

expandindo

0.65

transQ

0.64

 disambiguazione

0.61

Personendaten

0.60

Activations Density 0.343%

phrases that denote rankings or comparisons

No Comments

No Known Activations