INDEX

Explanations

terms related to programming and object structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

Ãªs

-0.16

ëĥ¥

-0.15

wort

-0.15

 serialVersionUID

-0.15

ç¹Ķ

-0.15

ump

-0.14

eldo

-0.14

 ÙĪØ§ÙĦØ¯

-0.14

lamaz

-0.14

łĢ

-0.14

POSITIVE LOGITS

 Nail

0.18

_PD

0.18

bers

0.16

Ol

0.15

ail

0.15

level

0.14

vo

0.14

pd

0.14

-by

0.14

 Bern

0.13

Activations Density 0.085%

terms related to programming and object structures

No Comments

No Known Activations