INDEX

Explanations

statements of denial and claims of innocence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

 sanctioned

-0.16

Ãºi

-0.15

Ð»Ð°Ñĩ

-0.15

 Sher

-0.15

quez

-0.15

ÑĢÐµÐ²

-0.14

xon

-0.14

OrDefault

-0.14

POSITIVE LOGITS

affles

0.15

oria

0.15

orus

0.14

ToEnd

0.14

atham

0.14

 against

0.14

acos

0.14

ivil

0.13

iox

0.13

aroo

0.13

Activations Density 0.160%

statements of denial and claims of innocence

No Comments

No Known Activations