INDEX

Explanations

words that indicate familial relationships or dynamics

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

é¾Ħ

-0.16

 Schmidt

-0.15

iaux

-0.15

-0.14

aliz

-0.14

æı´

-0.14

 evapor

-0.14

mek

-0.14

-0.13

 Trem

-0.13

POSITIVE LOGITS

dra

0.17

itel

0.16

iglia

0.15

inese

0.14

ulators

0.14

 Sister

0.14

 Monkey

0.14

uant

0.14

 meaning

0.13

 Craig

0.13

Activations Density 0.001%

words that indicate familial relationships or dynamics

No Comments

No Known Activations