INDEX

Explanations

terms related to depth and its variations in context

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

depth chart, depth charges, depth of field

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

WebServlet

-0.74

mobileqq

-0.54

RequestParam

-0.54

banners

-0.52

 noDo

-0.52

 unmarried

-0.51

__(/*!

-0.51

 Handels

-0.50

rules

-0.50

UserScript

-0.50

POSITIVE LOGITS

 Depth

1.21

 depth

1.16

Depth

1.06

depth

1.05

 DEPTH

0.96

DEPTH

0.88

 profundidad

0.77

 depths

0.76

 Tiefe

0.68

 profund

0.62

Activations Density 0.008%

terms related to depth and its variations in context

depth chart, depth charges, depth of field

No Comments

No Known Activations