INDEX

Explanations

references to various forms of documents and presentations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

aney

-0.16

etri

-0.15

Duy

-0.14

IDD

-0.13

opak

-0.13

loit

-0.13

 thÃ¢n

-0.13

udeau

-0.13

Goa

-0.13

wo

-0.13

POSITIVE LOGITS

instanc

0.16

usra

0.14

 threaten

0.14

|array

0.14

reich

0.13

 defaultManager

0.13

olleyError

0.13

:animated

0.13

_CRYPTO

0.13

 ÑģÑĦÐµÑĢ

0.13

Activations Density 0.152%

references to various forms of documents and presentations

No Comments

No Known Activations