INDEX

Explanations

phrases that emphasize significance or key concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

-0.15

burg

-0.14

iards

-0.14

 slightest

-0.13

ffects

-0.13

ÑĢÑĥ

-0.13

 sighting

-0.13

 whereabouts

-0.13

ould

-0.12

necessary

-0.12

POSITIVE LOGITS

 reason

0.42

 thing

0.41

 problem

0.36

 interesting

0.32

 Thing

0.30

 trouble

0.30

Thing

0.29

 funny

0.28

 beauty

0.28

 trick

0.28

Activations Density 0.344%

phrases that emphasize significance or key concepts

No Comments

No Known Activations