INDEX

Explanations

phrases and words related to highlighting or identifying important information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

ImageField

-0.68

 guts

-0.65

 SUDOC

-0.62

 McBride

-0.61

Dono

-0.61

 adipis

-0.59

leave

-0.59

odendron

-0.59

 embri

-0.58

pretation

-0.58

POSITIVE LOGITS

 highlights

1.00

 highlighting

0.98

 highlighted

0.96

 pointing

0.96

 points

0.96

Highlighted

0.93

 highlight

0.91

noted

0.90

Highlight

0.90

highlights

0.89

Activations Density 0.174%

phrases and words related to highlighting or identifying important information

No Comments

No Known Activations