INDEX

Explanations

assertions about perceptions or appearances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 Osprey

-0.48

 Colgate

-0.45

hyd

-0.45

 hosp

-0.43

 oxid

-0.42

 Livre

-0.42

 Hosp

-0.41

 manus

-0.41

OrderID

-0.40

 Holliday

-0.40

POSITIVE LOGITS

seemed

0.80

LookAnd

0.75

seem

0.74

 Seem

0.70

 seemed

0.70

 seem

0.70

seems

0.66

Seems

0.66

 seeming

0.63

 Seems

0.60

Activations Density 0.112%

assertions about perceptions or appearances

No Comments

No Known Activations