INDEX

Explanations

words associated with actions involving responsibility or oversight

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ena

-0.19

igkeit

-0.16

ach

-0.15

ritz

-0.15

itty

-0.14

lice

-0.14

ENA

-0.14

erras

-0.14

anco

-0.14

aw

-0.14

POSITIVE LOGITS

ing

0.19

à¤¼à¤°

0.16

/GPL

0.14

ä¸Ģä¸ĭ

0.14

 sÃ©

0.14

ExecutionContext

0.14

olta

0.14

CIT

0.14

 Bulk

0.14

bulk

0.14

Activations Density 0.248%

words associated with actions involving responsibility or oversight

No Comments

No Known Activations