INDEX

Explanations

mathematical notation and syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

htub

-0.15

elong

-0.14

ÐµÑĢÐº

-0.14

KIT

-0.14

 CascadeType

-0.14

åŃ

-0.14

erken

-0.14

æĺĩ

-0.14

annis

-0.14

XS

-0.14

POSITIVE LOGITS

 McCl

0.16

 racially

0.14

 mastur

0.14

0.13

inte

0.13

ea

0.13

 bitterness

0.13

id

0.13

Activations Density 0.076%

mathematical notation and syntax

No Comments

No Known Activations