INDEX

Explanations

terms related to safety and responsible practices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 kasarigan

-0.66

TypedDataSet

-0.61

RTLI

-0.58

siella

-0.58

InjectAttribute

-0.55

 كومونز

-0.55

Obrázky

-0.53

BeginContext

-0.52

 ArrayAdapter

-0.51

 GENERATED

-0.50

POSITIVE LOGITS

脚注の使い方

0.84

 safe

0.72

olesome

0.67

 toekomst

0.64

 secure

0.63

 safely

0.62

 sustainable

0.61

fortable

0.60

 Bale

0.60

 balanced

0.60

Activations Density 0.293%

terms related to safety and responsible practices

No Comments

No Known Activations