INDEX

Explanations

references to processes, progression, and results in varying contexts

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

Follows certain verbs/adjectives like created, containing, involved

np_acts-logits-general · gemini-2.0-flash

past participles followed by prepositions/related nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

:✨

-0.74

 незавершена

-0.74

parsedMessage

-0.66

 AssemblyCulture

-0.65

AutoModerator

-0.62

-------

-0.62

AnchorStyles

-0.61

⤹

-0.60

interopRequire

-0.60

MessageTagHelper

-0.59

POSITIVE LOGITS

السلام

0.38

 usage

0.36

ագրություններ

0.34

ControllerAdvice

0.34

 typelib

0.32

 PUSTAKA

0.32

 povezave

0.32

/";

0.31

/';

0.30

/,

0.30

Activations Density 1.559%

references to processes, progression, and results in varying contexts

Follows certain verbs/adjectives like created, containing, involved

past participles followed by prepositions/related nouns

No Comments

No Known Activations