INDEX

Explanations

occurrences of the word "chair" and references to hiding or showing elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

urlencoded

-0.43

MigrationBuilder

-0.39

覚えて

-0.39

Tikang

-0.38

multirow

-0.38

 Numerade

-0.37

ContextCompat

-0.37

 Beat

-0.36

 Sealed

-0.36

 Administrativna

-0.36

POSITIVE LOGITS

 hide

1.02

 Chair

0.93

Chair

0.92

 chair

0.92

 chairs

0.89

hide

0.89

 CHAIR

0.88

 Hide

0.88

chair

0.87

 Chairs

0.86

Activations Density 0.049%

occurrences of the word "chair" and references to hiding or showing elements

No Comments

No Known Activations