INDEX

Explanations

references to specific historical dates or periods

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

(le

-0.20

le

-0.19

les

-0.19

le

-0.18

 Leap

-0.17

 lease

-0.16

Les

-0.16

 precursor

-0.16

Les

-0.15

lea

-0.15

POSITIVE LOGITS

 partir

0.26

 travers

0.25

 prox

0.20

 titre

0.20

 propos

0.20

 cause

0.20

insi

0.19

 bord

0.18

 huis

0.18

 cÃ´tÃ©

0.18

Activations Density 0.010%

references to specific historical dates or periods

No Comments

No Known Activations