INDEX

Explanations

terms and phrases associated with espionage, piracy, and illicit activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 Walton

-0.17

mente

-0.16

eks

-0.15

alet

-0.15

éľŀ

-0.15

illes

-0.15

 hostage

-0.14

-bodied

-0.14

Eu

-0.14

POSITIVE LOGITS

ernet

0.18

ishly

0.16

ulence

0.15

buz

0.15

ê

0.14

Ø±ÙĪØ³

0.14

ous

0.14

ElementException

0.14

buster

0.14

ISM

0.14

Activations Density 0.187%

terms and phrases associated with espionage, piracy, and illicit activities

No Comments

No Known Activations