INDEX

Explanations

dialogue and conversational interaction

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

Ã´i

-0.14

aleb

-0.14

 eget

-0.14

Slo

-0.14

 Force

-0.13

bra

-0.13

alist

-0.13

å²Ĺ

-0.13

Force

-0.13

bilder

-0.13

POSITIVE LOGITS

 answer

0.47

 Answer

0.40

answer

0.38

 answered

0.37

 answers

0.36

åĽŀçŃĶ

0.36

 ANSW

0.35

 reply

0.34

çŃĶ

0.33

 answering

0.33

Activations Density 0.308%

dialogue and conversational interaction

No Comments

No Known Activations