INDEX

Explanations

terms related to toxicity and poisonous substances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 Kraw

-0.43

 amé

-0.42

CommonModule

-0.41

 classifying

-0.41

UnknownFieldSet

-0.40

bewerken

-0.40

 electrophoresis

-0.40

jsoup

-0.40

Cyfeiriadau

-0.39

 onBackPressed

-0.39

POSITIVE LOGITS

 toxic

0.75

Toxic

0.74

Pes

0.72

toxic

0.72

 Toxic

0.71

Pes

0.71

 TOXIC

0.70

 toxicity

0.68

 toxin

0.65

 toxins

0.65

Activations Density 0.419%

terms related to toxicity and poisonous substances

No Comments

No Known Activations