INDEX

Explanations

references to specific aviation incidents and casualties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

dil

-0.16

ayo

-0.16

iculo

-0.15

ÙĥØ±Ø©

-0.14

NON

-0.14

 cars

-0.14

icap

-0.14

anio

-0.14

rippling

-0.14

 Cooler

-0.14

POSITIVE LOGITS

 crash

0.24

 crashed

0.20

 Crash

0.20

 Flight

0.20

 crashes

0.18

olik

0.17

Flight

0.16

autop

0.16

flight

0.15

 crashing

0.15

Activations Density 0.028%

references to specific aviation incidents and casualties

No Comments

No Known Activations