INDEX

Explanations

terms related to topics being discussed or identified

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

Ŀ

-0.16

ouse

-0.16

 Ways

-0.15

eya

-0.15

acea

-0.15

lop

-0.15

orest

-0.14

bol

-0.14

ize

-0.14

 Mighty

-0.14

POSITIVE LOGITS

//{{

0.15

iaux

0.15

arella

0.15

erif

0.14

obox

0.14

abbo

0.14

æ°ĳæĹı

0.14

 Stock

0.14

tura

0.13

SES

0.13

Activations Density 0.009%

terms related to topics being discussed or identified

No Known Activations