INDEX

Explanations

terms related to toxicity, specifically in medical or scientific contexts

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

references to neurotoxicity and its effects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_mlp_out

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

 toxicity

-0.83

toxicity

-0.79

 Toxicity

-0.77

Toxicity

-0.69

Rptr

-0.65

 getRule

-0.65

toxic

-0.61

krankungen

-0.60

astify

-0.60

AddWithValue

-0.59

POSITIVE LOGITS

 pinulongan

0.84

 esternos

0.59

 المعيارى

0.53

UrlResolution

0.52

InputBorder

0.51

 AssemblyCompany

0.50

ateľ

0.50

 متعلقه

0.50

\\

0.50

تقاوى

0.50

Activations Density 0.157%

terms related to toxicity, specifically in medical or scientific contexts

references to neurotoxicity and its effects

No Comments

No Known Activations