INDEX

Explanations

頻繁に出現する一般的な名詞や接頭辞の使用。

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

insk

-0.17

uve

-0.16

insky

-0.16

zin

-0.15

elles

-0.15

alat

-0.15

-0.14

oa

-0.14

datas

-0.14

olicit

-0.14

POSITIVE LOGITS

htags

0.18

leftright

0.15

omial

0.15

Ð°ÑĤÐºÐ¾Ð²

0.15

 Bast

0.14

osten

0.14

bjerg

0.14

hdl

0.14

htag

0.14

ifest

0.14

Activations Density 0.002%

頻繁に出現する一般的な名詞や接頭辞の使用。

No Comments

No Known Activations