INDEX

Explanations

mathematical notations and structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

è¸ı

-0.15

elpers

-0.14

äºĭ

-0.13

 Ð¿ÑĢÐ¸Ñģ

-0.13

braco

-0.13

.Cast

-0.13

.VisualBasic

-0.13

ÑĢÐ¸ÑĤÑĤÑı

-0.13

linger

-0.13

avanaugh

-0.13

POSITIVE LOGITS

ledo

0.15

LETTE

0.14

 Alice

0.14

box

0.14

ine

0.14

åı·

0.14

 Bien

0.14

 massaggi

0.14

icap

0.14

Activations Density 0.139%

mathematical notations and structures

No Comments

No Known Activations