INDEX

Explanations

prepositions linked to multiple words and complex concepts

oai_token-act-pair · gemini-2.0-flash

tokens typically representing personal reflections or memories

np_token-act-pair-logits · gpt-4o-mini

Personal thoughts and feelings

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mind

-0.86

 heart

-0.86

 head

-0.84

 eyes

-0.82

 hands

-0.81

 ears

-0.71

 mouth

-0.71

 face

-0.68

heart

-0.68

head

-0.65

POSITIVE LOGITS

ViewImports

0.89

 your

0.89

 swoją

0.86

 Roskov

0.85

his

0.85

StoryboardSegue

0.84

 своем

0.82

 svůj

0.80

seu

0.79

 autorytatywna

0.79

Activations Density 1.713%

prepositions linked to multiple words and complex concepts

tokens typically representing personal reflections or memories

Personal thoughts and feelings

No Comments

No Known Activations