INDEX

Explanations

key concepts related to strategies, effects, and mechanisms within various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

Ð»Ð°Ð·

-0.17

Ø³ÙĪØ¨

-0.16

wargs

-0.15

abinet

-0.15

ETA

-0.14

orsk

-0.14

Ã¥de

-0.14

sted

-0.13

ä¸ĺ

-0.13

 Trot

-0.13

POSITIVE LOGITS

 such

0.18

such

0.15

ãĤ·ãĥªãĥ¼ãĤº

0.15

 possibilities

0.15

Avg

0.14

å½¹

0.14

rice

0.14

ver

0.14

dk

0.14

 heiÃŁ

0.14

Activations Density 0.185%

key concepts related to strategies, effects, and mechanisms within various contexts

No Comments

No Known Activations