INDEX

Explanations

references to numerical data and document structures

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

le

-0.17

agle

-0.16

aste

-0.15

bud

-0.15

Wi

-0.14

 Craw

-0.14

 sost

-0.14

ï¿½

-0.14

on

-0.14

POSITIVE LOGITS

ENTA

0.18

adil

0.16

.getSeconds

0.15

odyn

0.15

>NN

0.15

#ac

0.15

assium

0.15

gÄ±n

0.15

muz

0.15

Î¤Î±

0.14

Activations Density 0.075%

references to numerical data and document structures

No Comments

No Known Activations