INDEX

Explanations

references to thrones and succession in a historical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 Tunnel

-0.16

ÎµÏį

-0.15

 Toggle

-0.15

 complement

-0.15

acket

-0.15

Tuy

-0.15

 Tarif

-0.14

idge

-0.14

rupa

-0.14

raid

-0.14

POSITIVE LOGITS

 throne

0.49

th

0.46

thr

0.42

 thrown

0.40

 Throne

0.39

 Thrones

0.34

rones

0.34

 Ð¿ÑĢÐµÑģÑĤ

0.34

THR

0.32

thrown

0.31

Activations Density 0.054%

references to thrones and succession in a historical context

No Comments

No Known Activations