INDEX

Explanations

references to popular culture and entertainment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

æĺ¯æĪĳ

-0.18

è¹

-0.14

Ð¾ÑĤÑĥ

-0.14

aden

-0.13

 blackColor

-0.13

âĢĲ

-0.13

å¶

-0.13

 jenter

-0.13

iser

-0.13

aket

-0.12

POSITIVE LOGITS

 apparently

0.27

 Apparently

0.21

 pretty

0.20

Apparently

0.20

 somehow

0.19

 reportedly

0.19

ccording

0.18

 stuff

0.18

 according

0.18

 folks

0.18

Activations Density 0.853%

references to popular culture and entertainment

No Comments

No Known Activations