INDEX

Explanations

references to television series and movies, particularly those featuring notable actors

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

/lg

-0.18

.Network

-0.15

/maps

-0.15

/inet

-0.15

ä½³

-0.14

 bÃŃ

-0.14

ç¹ģ

-0.14

 Reactive

-0.14

Moy

-0.14

âĶĺ

-0.14

POSITIVE LOGITS

gh

0.17

 holiday

0.16

 direct

0.16

atel

0.16

 Christmas

0.15

tam

0.15

Lep

0.15

 bunk

0.15

XXX

0.15

Activations Density 0.080%

references to television series and movies, particularly those featuring notable actors

No Comments

No Known Activations