INDEX

Explanations

mathematical notation or expressions involving indices or subscripts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

yer

-0.17

retim

-0.16

Atoms

-0.15

odesk

-0.14

ella

-0.14

_FUN

-0.14

.hxx

-0.14

FLAG

-0.14

arshal

-0.14

ICI

-0.14

POSITIVE LOGITS

æŁľ

0.15

 caval

0.15

apy

0.14

Ð¾Ð»ÑĮÐ½Ð¾

0.13

ettel

0.13

eri

0.13

ÙħØ©

0.13

IRST

0.13

 blat

0.13

PRINTF

0.13

Activations Density 0.018%

mathematical notation or expressions involving indices or subscripts

No Comments

No Known Activations