INDEX

Explanations

references to depth or intensity, particularly concerning the word 'deep'

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

AccessorTable

-1.06

 propOrder

-1.05

RegistryLite

-1.02

antMatchers

-0.96

 Roskov

-0.94

 >=",

-0.93

 AssemblyCompany

-0.91

kháu

-0.91

 يتيمه

-0.90

urlpatterns

-0.89

POSITIVE LOGITS

 deep

1.68

 DEEP

1.64

 Deep

1.62

Deep

1.54

deep

1.49

DEEP

1.47

 depth

1.42

 depths

1.38

 deeper

1.33

 Depth

1.33

Activations Density 0.054%

references to depth or intensity, particularly concerning the word 'deep'

No Comments

No Known Activations