INDEX

Explanations

phrases indicating intention, meaning, or purpose

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

ChildScrollView

-0.85

ftagPool

-0.68

writeFieldEnd

-0.67

PreferredItem

-0.66

WebServlet

-0.65

 typelib

-0.62

 EClass

-0.62

expandindo

-0.62

RenderAtEndOf

-0.62

็จ

-0.61

POSITIVE LOGITS

 intent

0.84

 intending

0.84

 intended

0.83

 meant

0.83

 intends

0.81

 intend

0.81

 intentions

0.79

 intention

0.77

intent

0.75

 intención

0.71

Activations Density 0.483%

phrases indicating intention, meaning, or purpose

No Comments

No Known Activations