INDEX

Explanations

references to organizations or groups associated with political or military 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

Ab

-0.20

Aj

-0.19

 abstraction

-0.17

 Burst

-0.15

_asm

-0.15

assi

-0.15

AF

-0.15

:async

-0.14

illy

-0.14

 axis

-0.14

POSITIVE LOGITS

ob

0.40

ab

0.39

abs

0.36

tab

0.33

able

0.33

ables

0.29

lab

0.29

Ð°Ð±

0.28

ub

0.27

ib

0.27

Activations Density 0.035%

references to organizations or groups associated with political or military actions

No Comments

No Known Activations