INDEX

Explanations

instances of the word "covered" and related terms discussing coverage in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

berdayakan

-0.47

gás

-0.41

 prochains

-0.39



-0.39

 artikkelen

-0.38

 שוליים

-0.38

utilisons

-0.37

 usah

-0.36

 Traducción

-0.36

 seragam

-0.35

POSITIVE LOGITS

 covered

0.84

 Covered

0.82

Covered

0.76

covered

0.72

 domain

0.72

 Domain

0.64

 covering

0.61

 Covering

0.60

 coating

0.57

 pathway

0.57

Activations Density 0.168%

instances of the word "covered" and related terms discussing coverage in various contexts

No Comments

No Known Activations