INDEX

Explanations

words like "where", "when", "wherever", "whenever", or "one" in the context of someone talking about those topics.

oai_token-act-pair · gemini-2.0-flash

where/place

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_8/width_16k/average_l0_52

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.8.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Where

-0.74

 places

-0.74

 steder

-0.72

where

-0.71

 where

-0.70

 Where

-0.69

place

-0.68

Places

-0.68

 Places

-0.67

的地方

-0.67

POSITIVE LOGITS

oblastoma

0.60

 crossorigin

0.59

//

0.54

 المعيارى

0.54

meanor

0.54

lottesville

0.54

omaterial

0.54

aarrggbb

0.53

 FormBuilder

0.53

OGND

0.53

Activations Density 48.727%

words like "where", "when", "wherever", "whenever", or "one" in the context of someone talking about those topics.

where/place

No Comments

No Known Activations