INDEX

Explanations

terms related to solving issues or addressing problems, particularly in medical and social contexts

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

Fixing a problem

np_acts-logits-general · gemini-2.0-flash

resolve and overcome problems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 Signalez

-0.79

 OnTrigger

-0.62

 propOrder

-0.60

apimachinery

-0.59

 noDo

-0.57

ⓧ

-0.56

unhofer

-0.56

DockStyle

-0.56

 насељу

-0.55

 समीक्षक

-0.55

POSITIVE LOGITS

 problemet

0.65

 wrongs

0.65

 problems

0.62

excess

0.61

 efficacement

0.60

 excesses

0.59

 pesky

0.58

 проблемы

0.57

 overcrowding

0.57

 problemas

0.56

Activations Density 0.552%

terms related to solving issues or addressing problems, particularly in medical and social contexts

Fixing a problem

resolve and overcome problems

No Comments

No Known Activations