INDEX

Explanations

words associated with causation, particularly in healthcare and environmental contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.41.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

onAttach

-0.44

 appena

-0.44

amentals

-0.42

колко

-0.42

しっかり

-0.42

ammans

-0.41

Incoming

-0.39

 ролях

-0.38

 BaseActivity

-0.38

FTFY

-0.38

POSITIVE LOGITS

 havoc

0.87

 irreparable

0.77

 headaches

0.76

 untold

0.69

 confusion

0.67

 chaos

0.67

 consternation

0.64

 considerable

0.63

 serious

0.60

 significant

0.60

Activations Density 0.839%

words associated with causation, particularly in healthcare and environmental contexts

No Known Activations