INDEX

Explanations

words related to moral judgments and criticisms of actions

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

Adjective followed by "to"

np_acts-logits-general · gemini-2.0-flash

evaluative qualities or judgments

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

rawDesc

-0.76

awtextra

-0.66

 aDecoder

-0.65

Geplaatst

-0.65

 endforeach

-0.64

 DIOS

-0.60

windowFixed

-0.59

endwhile

-0.58

onucleotide

-0.58

 CURIAM

-0.58

POSITIVE LOGITS

to

0.84

ViewFeatures

0.70

 knowing

0.62

 faptul

0.56

Халык

0.52

 having

0.50

 that

0.48

 hearing

0.47

 να

0.47

ที่จะ

0.46

Activations Density 0.187%

words related to moral judgments and criticisms of actions

Adjective followed by "to"

evaluative qualities or judgments

No Comments

No Known Activations