INDEX

Explanations

phrases that emphasize or draw attention to key points or highlights in a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

(`/

-0.73

loem

-0.71

httphttps

-0.68

nteral

-0.66

getDoctrine

-0.64

Matteo

-0.63

Cup

-0.62

mosis

-0.62

ellees

-0.62

POSITIVE LOGITS

 highlight

2.19

 Highlight

2.16

 highlights

2.14

 Highlights

2.05

highlights

2.03

Highlights

1.96

Highlight

1.90

 highlighting

1.85

highlight

1.83

 highlighted

1.72

Activations Density 0.071%

phrases that emphasize or draw attention to key points or highlights in a text

No Comments

No Known Activations