INDEX

Explanations

phrases indicating evaluation or judgment of individuals and groups

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

evaluation or judgement

np_acts-logits-general · gemini-2.0-flash

deeming or considering someone/something

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

 ModelExpression

-0.67

tagHelperRunner

-0.64

CloseOperation

-0.61

 avowed

-0.60

Apparently

-0.59

NameInMap

-0.58

 absten

-0.58

 credited

-0.57

기도

-0.56

 Alleg

-0.56

POSITIVE LOGITS

 superior

0.73

0.66

“

0.64

worthy

0.62

 worthy

0.61

 inferior

0.59

 worth

0.59

 suitable

0.58

worth

0.58

0.56

Activations Density 0.663%

phrases indicating evaluation or judgment of individuals and groups

evaluation or judgement

deeming or considering someone/something

No Comments

No Known Activations