INDEX

Explanations

instances of editing or update references in a technical or formal context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ð°Ð¼Ð¿

-0.16

ile

-0.15

ancer

-0.15

å·±

-0.14

Ø§Ùħ

-0.14

iler

-0.14

amar

-0.14

-0.13

opus

-0.13

erno

-0.13

POSITIVE LOGITS

0.17

note

0.16

 çİ©

0.15

ohl

0.15

 Note

0.14

atorial

0.14

ROID

0.14

 note

0.14

-lnd

0.14

Note

0.13

Activations Density 0.038%

instances of editing or update references in a technical or formal context

No Comments

No Known Activations