INDEX

Explanations

references to teachers or educational experiences

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

 Mohamed

-0.16

Ð°ÑĢÑĤÐ°Ð¼

-0.16

bib

-0.15

hol

-0.15

 baptized

-0.15

idon

-0.14

Ã©ric

-0.14

abcdefghijkl

-0.14

ellen

-0.14

hol

-0.14

POSITIVE LOGITS

Han

0.28

Han

0.23

jur

0.23

Transmission

0.21

fi

0.21

 transmitted

0.21

Fi

0.20

 transmitting

0.20

 Transmission

0.20

 transmissions

0.19

Activations Density 0.065%

references to teachers or educational experiences

No Comments

No Known Activations