INDEX

Explanations

colons and other forms of punctuation indicating lists or references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

 èĢ

-0.15

/Branch

-0.15

pery

-0.14

-eyed

-0.14

æ·¡

-0.14

ÑĢÑĥÐ³

-0.14

ÐµÐ»Ð¸Ñĩ

-0.13

zdy

-0.13

Ùĭ

-0.13

ursed

-0.13

POSITIVE LOGITS

xt

0.18

 comm

0.14

os

0.14

olin

0.14

onz

0.13

itize

0.13

ospace

0.13

atty

0.13

untu

0.13

Activations Density 0.009%

colons and other forms of punctuation indicating lists or references

No Comments

No Known Activations