INDEX

Explanations

mentions of wars, particularly World Wars I and II

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

yun

-0.17

yyy

-0.16

 third

-0.15

 fourth

-0.15

yyyy

-0.14

xxxxxxxx

-0.14

yk

-0.13

yen

-0.13

ropic

-0.13

ê³

-0.13

POSITIVE LOGITS

II

0.24

Two

0.23

Two

0.19

ll

0.18

âħ

0.18

-era

0.18

One

0.18

lord

0.17

edor

0.17

lords

0.17

Activations Density 0.006%

mentions of wars, particularly World Wars I and II

No Comments

No Known Activations