INDEX

Explanations

URLs and references to YouTube content

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

YouTube videos

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

出版年

-0.54

tvguidetime

-0.51

 comuniques

-0.46

 Exactos

-0.45

#+#

-0.45

posedge

-0.45

TextAppearance

-0.45

tonode

-0.44

 consultato

-0.43

PageIndex

-0.43

POSITIVE LOGITS

 YouTube

0.71

 videos

0.68

YouTube

0.68

 youtube

0.61

 Youtube

0.59

 video

0.57

videos

0.57

Youtube

0.56

youtube

0.55

 vídeo

0.55

Activations Density 0.047%

URLs and references to YouTube content

YouTube videos

No Comments

No Known Activations