INDEX

Explanations

positive emotional expressions and sentiments

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

expressions of satisfaction and reliability regarding services

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

positive sentiment descriptors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.20.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

 ſever

-0.64

 myſelf

-0.63

 leaſt

-0.62

ябре

-0.60

withIdentifier

-0.59

 متعلقه

-0.59

 dafs

-0.58

 jScrollPane

-0.57

 poveznice

-0.56

 depositphotos

-0.56

POSITIVE LOGITS

 glad

1.44

 Glad

1.32

Glad

1.28

 nice

1.27

glad

1.21

nice

1.02

 Nice

1.00

Nice

0.96

 freut

0.94

 happy

0.93

Activations Density 0.077%

positive emotional expressions and sentiments

expressions of satisfaction and reliability regarding services

positive sentiment descriptors

No Comments

No Known Activations