INDEX

Explanations

references to specific locations and legal actions

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

MEA

-0.15

illi

-0.15

OCR

-0.15

illard

-0.15

 Relative

-0.14

ysa

-0.14

yi

-0.14

Relative

-0.14

krit

-0.13

omik

-0.13

POSITIVE LOGITS

åĩĮ

0.15

 consequence

0.15

ÚĨÛĮ

0.14

ulton

0.14

 strict

0.14

lal

0.14

-ons

0.13

rique

0.13

oger

0.13

elters

0.13

Activations Density 0.174%

references to specific locations and legal actions

No Comments

No Known Activations