INDEX

Explanations

concepts related to extraordinary or peculiar events that attract attention

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 plenty

-0.15

orda

-0.15

olib

-0.14

 Soon

-0.13

 VERY

-0.13

lÃ¤

-0.13

æľīä¸Ģ

-0.13

 Definitely

-0.13

erno

-0.13

POSITIVE LOGITS

so

0.79

å¦ĤæŃ¤

0.66

 à¤ĩà¤¤à¤¨

0.63

 such

0.61

è¿Ļä¹Ī

0.60

éĤ£ä¹Ī

0.59

 tÃ£o

0.57

 tolik

0.52

 sÃ¥

0.52

such

0.50

Activations Density 0.689%

concepts related to extraordinary or peculiar events that attract attention

No Comments

No Known Activations