INDEX

Explanations

terms that convey dependability and uniformity in various contexts

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

kuuta

-0.59

 Schult

-0.56

el

-0.56

 NavController

-0.55

pok

-0.54

 McCullough

-0.54

 INTERESAR

-0.54

ValueStyle

-0.53

 estimés

-0.52

 vultures

-0.51

POSITIVE LOGITS

 Consistency

1.15

 consistent

1.07

 consistency

1.07

 Consistent

1.06

 cohé

1.05

consistency

1.03

Consistency

1.03

consistent

0.99

 inconsistent

0.95

 cdti

0.95

Activations Density 0.128%

terms that convey dependability and uniformity in various contexts

No Comments

No Known Activations