INDEX

Explanations

queries related to comparisons and specifications in discussions

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

how many

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

InputBorder

-0.78

 fieldNum

-0.74

awtextra

-0.70

ſammen

-0.70

 configureStore

-0.68

parsedMessage

-0.68

 ſont

-0.67

AddTagHelper

-0.64

AccessorTable

-0.64

Spolja

-0.63

POSITIVE LOGITS

تقاوى

0.35

 test

0.29

NameInMap

0.29

prod

0.28

RE

0.28

0.27

 sufficient

0.27

ThroughAttribute

0.27

PROD

0.27

Activations Density 0.164%

queries related to comparisons and specifications in discussions

how many

No Comments

No Known Activations