INDEX

Explanations

references to time, particularly years

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

UserRole

-0.14

OKIE

-0.14

oder

-0.14

 Ð½Ð°ÑĩÐµ

-0.14

anes

-0.13

 æ©

-0.13

 Furn

-0.13

cee

-0.13

atÄ±r

-0.13

ekli

-0.13

POSITIVE LOGITS

 worked

0.26

worked

0.21

 served

0.20

 lavor

0.20

 Ã§alÄ±ÅŁ

0.19

 trabaj

0.19

 Ð±ÑĭÐ»

0.18

 ÑĢÐ°Ð±Ð¾ÑĤ

0.17

 byÅĤ

0.17

erve

0.16

Activations Density 0.029%

references to time, particularly years

No Comments

No Known Activations