INDEX

Explanations

mentions of locations and references to collaborative projects or activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

eed

-0.15

nut

-0.14

 Fucked

-0.14

Ð½Ð°ÑĤ

-0.13

 fucked

-0.13

 menstr

-0.13

Ã¡i

-0.13

Ã©ment

-0.13

Äĵ

-0.13

fuck

-0.13

POSITIVE LOGITS

Sen

0.24

sen

0.24

 Just

0.22

 seniors

0.21

 Christian

0.20

 cover

0.20

SEN

0.19

 playing

0.19

Sen

0.19

 Henderson

0.18

Activations Density 0.000%

mentions of locations and references to collaborative projects or activities

No Comments

No Known Activations