INDEX

Explanations

specific characters or symbols that may indicate formatting or non-standard text elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

bach

-0.16

unge

-0.15

ungan

-0.15

 éķ

-0.13

 Vladim

-0.13

 ÑĢÐµÐ°Ðº

-0.13

.throw

-0.12

 ÐºÐ°Ð»ÑĮ

-0.12

 DeepCopy

-0.12

/Math

-0.12

POSITIVE LOGITS

 fire

0.27

 burner

0.25

 Fire

0.24

/fire

0.23

Fire

0.22

 burn

0.22

fire

0.22

çĩĥ

0.21

 burns

0.21

 flames

0.21

Activations Density 0.010%

specific characters or symbols that may indicate formatting or non-standard text elements

No Comments

No Known Activations