INDEX

Explanations

phrases that provide insights or understanding into various processes

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Paglinawan

-0.85

RegistryLite

-0.73

abetes

-0.73

TagHelper

-0.71

FunctionFlags

-0.70

dfrac

-0.67

 namanya

-0.67

bootstrapcdn

-0.67

angered

-0.66

batore

-0.65

POSITIVE LOGITS

 insight

1.54

 Insight

1.53

Insight

1.47

insight

1.34

 insights

1.28

 Insights

1.20

Insights

1.20

insights

1.19

 insightful

0.84

 Einblick

0.80

Activations Density 0.005%

phrases that provide insights or understanding into various processes

No Comments

No Known Activations