INDEX

Explanations

language related to rules, conditions, or restrictions

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

ussen

-0.17

avor

-0.17

ntax

-0.16

aned

-0.15

ilion

-0.15

-addons

-0.15

edith

-0.14

ABEL

-0.14

UNU

-0.14

oppins

-0.14

POSITIVE LOGITS

 Nack

0.17

.toolbox

0.15

 acad

0.14

ikit

0.14

 Galaxy

0.14

 Dwight

0.14

 Keller

0.14

ISM

0.14

 spotlight

0.13

rette

0.13

Activations Density 0.002%

language related to rules, conditions, or restrictions

No Comments

No Known Activations