INDEX

Explanations

expressions of strong affection or admiration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

lep

-0.17

afari

-0.15

bare

-0.15

eries

-0.14

Identity

-0.14

suz

-0.14

 Kale

-0.14

forces

-0.14

alles

-0.13

Insets

-0.13

POSITIVE LOGITS

OLON

0.17

iser

0.16

Verdana

0.14

antan

0.14

eer

0.14

isÃŃ

0.14

ãĤ¿ãĥ³

0.14

NodeType

0.14

 ÑĢÐ°Ð·Ð²Ð¸

0.13

conc

0.13

Activations Density 0.131%

expressions of strong affection or admiration

No Comments

No Known Activations