INDEX

Explanations

terms related to espionage and intelligence activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

anch

-0.17

roc

-0.16

ivol

-0.15

ointment

-0.15

Ø§Ø¦ÙĦ

-0.15

oha

-0.15

ppard

-0.15

è¨´

-0.14

æ´²

-0.14

iment

-0.14

POSITIVE LOGITS

 agent

0.17

-agent

0.15

 Burr

0.15

 máºŃt

0.15

agent

0.15

 agents

0.15

0.14

ighth

0.14

 Agent

0.14

gable

0.14

Activations Density 0.088%

terms related to espionage and intelligence activities

No Comments

No Known Activations