INDEX

Explanations

the pronoun "I" in various contexts

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

I

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

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

 autorytatywna

-1.21



-0.92

Tembelea

-0.90

 ProtoMessage

-0.85

aarrggbb

-0.82

OGND

-0.80

ChildScrollView

-0.80

+#+

-0.79

 ExecuteAsync

-0.78

 Roskov

-0.78

POSITIVE LOGITS

The

0.63

0.57

We

0.55

It

0.54

Is

0.52

In

0.52

is

0.51

There

0.50

If

0.49

Activations Density 0.075%

the pronoun "I" in various contexts

I

No Comments

No Known Activations