INDEX

Explanations

occurrences of the words "being" and "was" along with references to people or groups experiencing mistreatment or scrutiny

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

venture

-0.17

WXYZ

-0.16

Ã¡ce

-0.16

engkap

-0.16

loub

-0.16

@Web

-0.15

IBC

-0.15

ipay

-0.15

.ascii

-0.14

éŁ¿

-0.14

POSITIVE LOGITS

 targeted

0.30

 attacked

0.27

amb

0.26

 subjected

0.24

 targets

0.24

 approached

0.24

 bomb

0.24

hit

0.24

 subject

0.23

 barr

0.22

Activations Density 0.214%

occurrences of the words "being" and "was" along with references to people or groups experiencing mistreatment or scrutiny

No Comments

No Known Activations