INDEX

Explanations

assertions and denials in discussions about evidence or claims

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/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.35.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

 ModelExpression

-0.47

Opis

-0.45

AxisSize

-0.34

模様

-0.34

 attendu

-0.33

 palanca

-0.33

+#+#

-0.33

 Opis

-0.33

SIGNAL

-0.33

 porcelaine

-0.32

POSITIVE LOGITS

 refute

1.70

 refuted

1.63

 disprove

1.50

 rebuttal

1.46

 rebut

1.40

deb

1.35

 contradicted

1.34

 counter

1.30

 contradict

1.27

 contradicts

1.26

Activations Density 0.948%

assertions and denials in discussions about evidence or claims

No Comments

No Known Activations