INDEX

Explanations

references to design and architecture

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

ouston

-0.18

ht

-0.16

znÃ¡m

-0.15

ynch

-0.15

verture

-0.15

th

-0.15

hd

-0.14

ursal

-0.14

bes

-0.14

onya

-0.14

POSITIVE LOGITS

akis

0.17

erset

0.16

å²Ĺ

0.16

à¸¥à¸²à¸¢

0.15

offs

0.15

 Ã¼r

0.15

kont

0.14

anne

0.14

áº¿u

0.14

Ð¾Ð²ÑĭÐ¹

0.14

Activations Density 0.037%

references to design and architecture

No Comments

No Known Activations