INDEX

Explanations

references to the word "the" in a variety of contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

 helyzet

-0.50

 wezen

-0.45

 المعيارى

-0.45

ThroughAttribute

-0.42

 mãi

-0.41

 skoro

-0.40

 Exactos

-0.39

mapStateToProps

-0.39

 másik

-0.38

 ferons

-0.38

POSITIVE LOGITS

 occasional

0.71

 equivalent

0.66

 same

0.66

 finest

0.66

 fewest

0.64

 smallest

0.63

 rarest

0.61

 usual

0.61

occas

0.61

 safest

0.60

Activations Density 1.302%

references to the word "the" in a variety of contexts

No Comments

No Known Activations