INDEX

Explanations

phrases related to personal identity and self-introduction

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

the neuron detects words and phrases expressing intention, willingness, ability, or deliberate action (volition).

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 потре

-0.35

 AttributeSet

-0.33

 Luft

-0.33

$\

-0.31

ませんでした

-0.31

$\

-0.30

 poved

-0.30

 fitting

-0.30

 fully

-0.30

↵↵

-0.29

POSITIVE LOGITS

ſelf

0.86

ſelves

0.77

 ſche

0.68

aarrggbb

0.67

 ſta

0.65

 myſelf

0.64

 houſe

0.63

 pleaſure

0.62

 Houſe

0.62

<unused14>

0.61

Activations Density 0.168%

phrases related to personal identity and self-introduction

the neuron detects words and phrases expressing intention, willingness, ability, or deliberate action (volition).

No Comments

No Known Activations