INDEX

Explanations

structured information and contextual details about a subject

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

ValueStyle

-0.72

Efq

-0.69

 houſe

-0.69

 Reſ

-0.69

 Monfieur

-0.68

 cauſe

-0.65

Abit

-0.65

 purpoſe

-0.64

Попис

-0.63

 Diſ

-0.63

POSITIVE LOGITS

 note

0.76

 extra

0.72

 Note

0.68

add

0.67

 notes

0.67

 mention

0.65

notes

0.64

 Notes

0.64

 Additional

0.62

 supplemental

0.62

Activations Density 0.535%

structured information and contextual details about a subject

No Comments

No Known Activations