INDEX

Explanations

descriptive terms related to horror and extreme negative experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

tableFuture

-0.75

stateProvider

-0.75

 فريبيس

-0.72

 Discre

-0.67

AccessorTable

-0.66

providedIn

-0.66

createCanvas

-0.66

AddTagHelper

-0.64

MENAFN

-0.63

ferrer

-0.63

POSITIVE LOGITS

 horrible

1.30

 terrible

1.07

 disgusting

1.05

Horrible

1.04

 hideous

1.03

 awful

1.02

 horrific

1.00

 horrid

1.00

 horrifying

0.97

 horror

0.96

Activations Density 0.238%

descriptive terms related to horror and extreme negative experiences

No Comments

No Known Activations