INDEX

Explanations

specific references to posts and articles, especially in a blog or technical context

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

Tokens appearing after the word "post"

np_acts-logits-general · gemini-2.0-flash

view this post

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.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

MigrationBuilder

-0.60

 CreateTagHelper

-0.48

 those

-0.48

 lenker

-0.46

MessageOf

-0.46

 dabei

-0.44

 contextLoads

-0.43

 beginnetje

-0.42

 therein

-0.42

 saites

-0.42

POSITIVE LOGITS

 PARTICULAR

0.56

particular

0.55

 particular

0.55

PARTIC

0.53

 alone

0.52

 particolare

0.51

Particular

0.51

Espèce

0.50

 ALONE

0.50

 particulares

0.48

Activations Density 0.748%

specific references to posts and articles, especially in a blog or technical context

Tokens appearing after the word "post"

view this post

No Comments

No Known Activations