INDEX

Explanations

references to statistical or numerical data, particularly in a tabular or structured format

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

AddTagHelper

-0.60

IntoConstraints

-0.57

OGND

-0.57

VIAF

-0.47

 surla

-0.46

ThroughAttribute

-0.45

 kaarangay

-0.44

zzleHttp

-0.43

 pleaſure

-0.43

intenant

-0.41

POSITIVE LOGITS

EndContext

0.40

 Administrativna

0.39

 çiz

0.38

 detailsLots

0.38

 counselors

0.37

 Constitucional

0.37

مقاله

0.36

StateToProps

0.35

ếng

0.35

WriteAttribute

0.35

Activations Density 0.122%

references to statistical or numerical data, particularly in a tabular or structured format

No Comments

No Known Activations