INDEX

Explanations

references to various forms of toxicity

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 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

ArgumentParser

-0.97

 Paglinawan

-0.96

 iſt

-0.90

CppMethod

-0.90

Билгалдахарш

-0.89

berdayakan

-0.89

webElementXpaths

-0.88

Identyfik

-0.87



-0.87

tagHelperRunner

-0.86

POSITIVE LOGITS

[toxicity=0]

3.49

↵↵

1.14

↵

0.81

</h6>

0.80

0.79

toxicity

0.70

0.69

//

0.67

<eos>

0.66

Activations Density 0.052%

references to various forms of toxicity

No Comments

No Known Activations