INDEX

Explanations

terms and phrases that indicate unusual or bizarre qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

IsContent

-0.74

 ComVisible

-0.70

multicolumn

-0.68

 coals

-0.67

CardBody

-0.65

شرين

-0.65

 Audiodateien

-0.62

WritableDatabase

-0.61

WriteLiteral

-0.60

PreAuthorize

-0.60

POSITIVE LOGITS

 strange

1.21

 Strange

1.19

strange

1.13

 weird

1.09

 wierd

1.05

 bizarre

1.05

 Weird

1.04

Strange

1.02

weird

1.00

Weird

0.98

Activations Density 0.129%

terms and phrases that indicate unusual or bizarre qualities

No Comments

No Known Activations