INDEX

Explanations

references to specific geographical locations and measurements

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

a direction or proximity

np_acts-logits-general · gemini-2.0-flash

location descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

SharedCtor

-0.73

 intptr

-0.70

#+#

-0.65

 EconPapers

-0.63

SequentialGroup

-0.57

voyance

-0.56

protoimpl

-0.55

ousness

-0.54

skjaer

-0.54

 mergeFrom

-0.53

POSITIVE LOGITS

 near

0.99

near

0.85

Near

0.74

 towards

0.71

 dekat

0.69

in

0.68

 toward

0.66

 somewhere

0.66

towards

0.65

toward

0.63

Activations Density 0.351%

references to specific geographical locations and measurements

a direction or proximity

location descriptions

No Comments

No Known Activations