INDEX

Explanations

phrases that involve relationships or dynamics between entities, particularly focusing on historical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

 frattempo

-0.50

 SwitchCompat

-0.49

 Schließlich

-0.47

 finalmente

-0.47

 finally

-0.46

finally

-0.46

principalColumn

-0.45

 zwar

-0.45

 MainAxisSize

-0.45

 lastly

-0.45

POSITIVE LOGITS

 beginning

0.94

 inception

0.93

 outset

0.84

beginning

0.82

 very

0.81

 BEGINNING

0.78

一开始

0.69

 conception

0.69

 Beginning

0.68

Beginning

0.68

Activations Density 0.183%

phrases that involve relationships or dynamics between entities, particularly focusing on historical context

No Comments

No Known Activations