INDEX

Explanations

instances of specific characters or symbols often found in digital text formats

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

ustum

-0.15

å°Ĭ

-0.14

 pornos

-0.14

enek

-0.14

 Ð·Ð¾Ð²

-0.14

oltip

-0.13

à¹Ħà¸¥à¸Ļ

-0.13

ongan

-0.13

fragment

-0.13

Î»Î»Î·

-0.13

POSITIVE LOGITS

 feeling

0.25

 fearing

0.23

 Feeling

0.22

 anxiety

0.20

 believing

0.20

 worry

0.19

fix

0.19

 paranoia

0.19

 thinking

0.19

 worried

0.18

Activations Density 0.033%

instances of specific characters or symbols often found in digital text formats

No Comments

No Known Activations