INDEX

Explanations

instances of numbers, counting, and references to quantities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

leh

-0.14

orelease

-0.14

NSIndexPath

-0.14

emann

-0.14

enderit

-0.14

RectTransform

-0.13

amat

-0.13

.createTextNode

-0.13

ajes

-0.13

 Ending

-0.13

POSITIVE LOGITS

 began

0.69

 begin

0.68

 begins

0.63

 begun

0.62

begin

0.61

å¼Ģå§ĭ

0.60

 beginning

0.58

 started

0.56

 starts

0.54

éĸĭå§ĭ

0.54

Activations Density 0.583%

instances of numbers, counting, and references to quantities

No Comments

No Known Activations