INDEX

Explanations

references to mortality and death-related events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

éļ

-0.18

ernes

-0.17

annah

-0.17

odp

-0.15

å¬

-0.15

ibu

-0.14

 muse

-0.14

oday

-0.14

olves

-0.14

 outgoing

-0.14

POSITIVE LOGITS

atorium

0.15

OPY

0.14

atoria

0.14

rimon

0.14

igar

0.14

 shall

0.14

 Challenger

0.14

strup

0.14

åĿ¡

0.13

asl

0.13

Activations Density 0.074%

references to mortality and death-related events

No Comments

No Known Activations