INDEX

Explanations

names of individuals and family relationships

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

esub

-0.18

ÐµÐ·

-0.16

olla

-0.15

INET

-0.15

esen

-0.14

 sana

-0.14

jew

-0.14

CreateInfo

-0.14

>Returns

-0.14

(DBG

-0.14

POSITIVE LOGITS

ault

0.17

0.15

 Wolfe

0.15

uell

0.14

udit

0.14

Ø±ÙģØª

0.14

 Byron

0.14

Î±Î½Î´

0.14

uese

0.14

 Buying

0.13

Activations Density 0.008%

names of individuals and family relationships

No Comments

No Known Activations