INDEX

Explanations

references to specific locations and their attributes

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

"[something] is located"

np_acts-logits-general · gemini-2.0-flash

location after is/are/was

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

paramref

-0.32

ﬅ

-0.29

省市镇

-0.29

 saurait

-0.28

żesz

-0.28

ृ

-0.27

 auxquels

-0.27

 stronę

-0.27

inglés

-0.26

[:,:,

-0.26

POSITIVE LOGITS

 located

2.78

located

2.45

 Located

2.38

Located

2.33

 situated

2.31

 ubicado

1.99

 ubicada

1.91

 Situated

1.88

 situé

1.86

 terletak

1.84

Activations Density 0.651%

references to specific locations and their attributes

"[something] is located"

location after is/are/was

No Comments

No Known Activations