INDEX

Explanations

references to people, places, or entities related to combined concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

cola

-0.17

unas

-0.17

Ä±da

-0.14

 heroes

-0.14

execute

-0.13

à¹Ĭ

-0.13

ontent

-0.13

Ø¹Ø§ÙĨ

-0.13

/stdc

-0.13

antage

-0.13

POSITIVE LOGITS

/or

0.22

nbsp

0.21

iesen

0.17

quot

0.17

raquo

0.17

rea

0.17

ahl

0.17

 alike

0.16

/of

0.16

 Others

0.15

Activations Density 0.273%

references to people, places, or entities related to combined concepts

No Comments

No Known Activations