INDEX

Explanations

actions related to maintenance and repair

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

atto

-0.15

idon

-0.15

ianne

-0.14

uluk

-0.14

dag

-0.14

StateChanged

-0.14

æ¼ı

-0.14

Closure

-0.14

ICON

-0.14

illi

-0.14

POSITIVE LOGITS

 removal

0.31

 Removal

0.30

 remove

0.28

 removing

0.27

 Remove

0.26

 Removing

0.26

Removing

0.25

remove

0.25

-remove

0.25

 remover

0.24

Activations Density 0.078%

actions related to maintenance and repair

No Comments

No Known Activations