INDEX

Explanations

references to bees and beekeeping

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

amura

-0.16

cret

-0.16

ãģĤãģ£ãģŁ

-0.16

tt

-0.15

stit

-0.15

ecko

-0.15

stad

-0.15

ynos

-0.15

sten

-0.15

agher

-0.15

POSITIVE LOGITS

hive

0.29

keeping

0.28

keepers

0.28

keeper

0.27

 hive

0.25

keep

0.22

èľľ

0.21

indr

0.21

 Hive

0.21

 keeper

0.20

Activations Density 0.006%

references to bees and beekeeping

No Comments

No Known Activations