INDEX

Explanations

phrases indicating need, access, or functionality in technical contexts

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ẻo

-0.34

操

-0.32

休

-0.30

Ger

-0.30

web

-0.29

imp

-0.28

ManyToOne

-0.28

sp

-0.27

██

-0.27

Sug

-0.27

POSITIVE LOGITS

AddTagHelper

0.85

AndEndTag

0.82

PreferredItem

0.77

 فريبيس

0.71

 autorytatywna

0.69

 ddelweddau

0.69

AnimationsModule

0.68

TintMode

0.68

évaluateur

0.68

 Infórmanos

0.67

Activations Density 0.058%

phrases indicating need, access, or functionality in technical contexts

No Comments

No Known Activations