INDEX

Explanations

references to toxicity in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 increí

-0.59

idoo

-0.58

 incrí

-0.56

򐂰

-0.56

 stiefe

-0.55

liesslich

-0.55

IndentedString

-0.53

 beſch

-0.52

})->

-0.52

 Airborne

-0.51

POSITIVE LOGITS

toxicity

2.56

 toxicity

0.89

toxic

0.76

Toxicity

0.75

 Toxicity

0.73

xicity

0.71

TOXIC

0.59

 TOXIC

0.56

 toxic

0.54

httphttps

0.52

Activations Density 0.041%

references to toxicity in various contexts

No Comments

No Known Activations