INDEX

Explanations

conjunctions and cause-and-effect relationships in the text

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

UnusedPrivate

-0.75

 المعيارى

-0.73

 дописавши

-0.71

 GenerationType

-0.69

WriteTagHelper

-0.69

पया

-0.68

IsMutable

-0.66

蚪

-0.65

UserScript

-0.64

#+#

-0.64

POSITIVE LOGITS

 although

0.56

 since

0.50

 they

0.50

 gdyż

0.47

Although

0.47

Since

0.47

although

0.47

since

0.45

 obwohl

0.45

 chociaż

0.43

Activations Density 0.375%

conjunctions and cause-and-effect relationships in the text

No Comments

No Known Activations