INDEX

Explanations

numeric values and related data points

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

fold

-0.23

ths

-0.20

rd

-0.18

th

-0.17

erman

-0.17

one

-0.16

Ãłng

-0.16

rid

-0.15

iper

-0.15

ikal

-0.15

POSITIVE LOGITS

st

0.23

Password

0.20

 liners

0.19

MDB

0.18

 thing

0.17

 Corinthians

0.17

/XMLSchema

0.17

 hour

0.16

Ð¾Ð´ÐµÑĢÐ¶

0.16

drv

0.16

Activations Density 0.115%

numeric values and related data points

No Comments

No Known Activations