INDEX

Explanations

phrases questioning the credibility or validity of information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

urrets

-0.17

Ð½Ð¸ÑĨ

-0.16

Ð²Ð¸ÑĤ

-0.15

 Conveyor

-0.15

Ctrls

-0.14

 CreateTable

-0.14

oley

-0.14

_wc

-0.14

roman

-0.14

atik

-0.14

POSITIVE LOGITS

 satellite

0.32

 Satellite

0.30

 Free

0.27

 Dish

0.27

 dish

0.26

 satellites

0.25

 reception

0.25

 dishes

0.24

Sky

0.24

 bouquet

0.23

Activations Density 0.037%

phrases questioning the credibility or validity of information

No Comments

No Known Activations