INDEX

Explanations

references to historical figures and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

afil

-0.16

 èĹ¤

-0.15

ussen

-0.15

iaux

-0.15

chl

-0.15

çŃĴ

-0.14

Ø§Ø¯Øª

-0.14

 chute

-0.14

Î¼ÏĨ

-0.14

borg

-0.14

POSITIVE LOGITS

 convict

0.32

 Governor

0.26

 conv

0.23

 Conv

0.23

0.22

conv

0.22

Conv

0.21

 Colony

0.21

0.20

 governor

0.20

Activations Density 0.030%

references to historical figures and places

No Comments

No Known Activations