INDEX

Explanations

textual references to opposing parties or positions in arguments

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

 Plenum

-0.51

 uLocal

-0.46

Btu

-0.43

saraba

-0.42

uação

-0.42

setCellValue

-0.41

 viață

-0.41

chymal

-0.41

 שוליים

-0.40

ulemon

-0.40

POSITIVE LOGITS

 against

1.20

 Against

1.20

Against

1.17

against

1.13

 AGAINST

1.13

 contre

0.81

 gegen

0.79

 melawan

0.75

 נגד

0.71

 против

0.66

Activations Density 0.243%

textual references to opposing parties or positions in arguments

No Comments

No Known Activations