INDEX

Explanations

proper nouns, particularly names and titles related to various figures and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

isman

-0.15

idel

-0.15

ÐºÑĢÐ°

-0.15

ait

-0.14

uv

-0.14

ibox

-0.14

hik

-0.14

 Provid

-0.14

weather

-0.14

asio

-0.14

POSITIVE LOGITS

Ä¾

0.15

-relative

0.15

 gratuiti

0.14

.svg

0.14

.createFrom

0.14

ignKey

0.14

ipi

0.14

 polo

0.14

tier

0.13

antine

0.13

Activations Density 0.026%

proper nouns, particularly names and titles related to various figures and entities

No Comments

No Known Activations