INDEX

Explanations

academic affiliations and professional titles within educational or research institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 Ø¨Ø§ÙĤ

-0.16

OUND

-0.15

ihn

-0.15

azzi

-0.14

esto

-0.14

 muster

-0.14

ibr

-0.14

ç¸

-0.14

FOUNDATION

-0.14

ÅĤo

-0.14

POSITIVE LOGITS

 prior

0.18

 awards

0.15

uge

0.15

thr

0.15

 interests

0.15

Ð³Ð¾Ð´

0.15

 Prior

0.15

(with

0.14

Prior

0.14

 apart

0.14

Activations Density 0.047%

academic affiliations and professional titles within educational or research institutions

No Comments

No Known Activations