INDEX

Explanations

references to educational and legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

:description

-0.15

VK

-0.14

inst

-0.14

ickle

-0.14

ephy

-0.14

Yug

-0.14

cona

-0.13

asaki

-0.13

ipt

-0.13

itzer

-0.13

POSITIVE LOGITS

ÑĨÑİ

0.15

Ã¤ch

0.15

ickey

0.15

 voks

0.14

defaultValue

0.14

rive

0.14

Mutable

0.14

TextStyle

0.14

 Lovely

0.14

stal

0.14

Activations Density 0.753%

references to educational and legal contexts

No Comments

No Known Activations