INDEX

Explanations

specific punctuations and formatting in text

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

flen

-0.17

abay

-0.16

ilst

-0.15

fds

-0.15

cth

-0.15

 ÙħÛĮÙĦ

-0.15

contrast

-0.14

acific

-0.14

edir

-0.14

foy

-0.14

POSITIVE LOGITS

ota

0.17

orf

0.16

 eslint

0.16

steam

0.15

à¸Ĥà¸§

0.15

Î½Î¹

0.15

 loved

0.15

Î¹Î±Îº

0.14

atan

0.14

Activations Density 0.001%

specific punctuations and formatting in text

No Comments

No Known Activations