INDEX

Explanations

themes of being trapped or involved in conspiracies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

Î´Î®

-0.15

oppers

-0.15

riad

-0.15

uchi

-0.14

endir

-0.14

è¾°

-0.14

_REGISTRY

-0.13

å±Ĭ

-0.13

ataka

-0.13

 sond

-0.13

POSITIVE LOGITS

 caught

0.31

caught

0.27

 Caught

0.25

 involved

0.24

Caught

0.22

 catch

0.21

catch

0.21

 involvement

0.20

 catches

0.20

åį·

0.18

Activations Density 0.107%

themes of being trapped or involved in conspiracies

No Comments

No Known Activations