INDEX

Explanations

expressions and symbols related to mathematical definitions and limits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

er

-0.18

 Prev

-0.16

Er

-0.15

 subs

-0.15

rac

-0.15

osal

-0.15

Er

-0.14

 Barker

-0.14

Enter

-0.14

Fut

-0.14

POSITIVE LOGITS

 infinity

0.38

infinity

0.35

 Infinity

0.33

.infinity

0.32

Infinity

0.32

 infinite

0.27

 Infinite

0.26

_INFINITY

0.24

.inf

0.23

Inf

0.22

Activations Density 0.104%

expressions and symbols related to mathematical definitions and limits

No Comments

No Known Activations