INDEX

Explanations

terminology related to laws and rules

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Hakk

-0.18

avian

-0.16

hythm

-0.15

iem

-0.14

 Gott

-0.14

oen

-0.14

arial

-0.14

ista

-0.14

Ø§ÙģÙĤ

-0.14

POSITIVE LOGITS

osed

0.16

importe

0.15

uters

0.14

'gc

0.14

cplusplus

0.14

Ú©Ø²

0.13

ties

0.13

 bear

0.13

gran

0.13

alia

0.13

Activations Density 0.014%

terminology related to laws and rules

No Comments

No Known Activations