INDEX

Explanations

phrases indicating consequences or liabilities related to actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

strup

-0.17

ynes

-0.15

 Ledger

-0.14

uben

-0.14

inya

-0.14

lfw

-0.14

irit

-0.14

Äĥn

-0.14

-0.13

POSITIVE LOGITS

 risk

0.44

risk

0.39

 Risk

0.37

 risks

0.37

Risk

0.34

 risking

0.31

-risk

0.30

Ris

0.28

ris

0.27

 runs

0.25

Activations Density 0.209%

phrases indicating consequences or liabilities related to actions

No Comments

No Known Activations