INDEX

Explanations

terms related to wildfires and evacuation efforts

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

before medical or fire-related terms

np_acts-logits-general · gemini-2.0-flash

malignant" or "margins" or "fire"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 intptr

-0.45

-0.40

tonsoft

-0.39

 TextInputType

-0.39

dol

-0.38

 mises

-0.37

 ECONOM

-0.36

Dol

-0.36

DataBinding

-0.35

 trä

-0.35

POSITIVE LOGITS

 CreateTagHelper

0.59

tableFuture

0.52

 wildfire

0.52

 melanoma

0.50

 GenerationType

0.48

onOptions

0.47

mybatisplus

0.47

PMailer

0.43

 nakalista

0.41

OGND

0.40

Activations Density 0.334%

terms related to wildfires and evacuation efforts

before medical or fire-related terms

malignant" or "margins" or "fire"

No Comments

No Known Activations