INDEX

Explanations

names and terms related to authors and notable individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

'gc

-0.19

ubs

-0.17

 tinh

-0.15

ÃŃsto

-0.14

à¥ģà¤Ĺà¤¤

-0.14

 ØªØ¹Ø¯

-0.14

irmed

-0.14

æĺĩ

-0.13

Ð²Ð¸ÑĤ

-0.13

rics

-0.13

POSITIVE LOGITS

łí

0.16

(Void

0.15

echa

0.15

izzo

0.14

awy

0.14

à¥įà¤ľ

0.14

hawk

0.13

Äĥn

0.13

alled

0.13

otron

0.13

Activations Density 0.108%

names and terms related to authors and notable individuals

No Comments

No Known Activations