INDEX

Explanations

conjunctions and connective phrases that indicate relationships between ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

TagHelper

-0.65

URLConnection

-0.54

acterial

-0.53

sitive

-0.52

vábbi

-0.52

lotz

-0.52

 betekenis

-0.51

QtWidgets

-0.51

 germs

-0.51

再说

-0.51

POSITIVE LOGITS

 must

0.93

can

0.90

 will

0.89

is

0.88

has

0.81

was

0.80

may

0.78

 merupakan

0.75

 should

0.67

 represents

0.66

Activations Density 0.368%

conjunctions and connective phrases that indicate relationships between ideas

No Comments

No Known Activations