INDEX

Explanations

references to time periods and lengths of effort

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

:CGRect

-0.18

ahlen

-0.16

ÏĦÎ¯

-0.15

aldi

-0.15

PerPixel

-0.15

coverage

-0.14

EventManager

-0.14

vvm

-0.14

AspNet

-0.13

 coverage

-0.13

POSITIVE LOGITS

 searching

0.14

iano

0.14

 effort

0.14

cul

0.14

ultiple

0.13

ated

0.13

 Magical

0.13

FindObject

0.13

 quiet

0.13

xor

0.13

Activations Density 0.149%

references to time periods and lengths of effort

No Comments

No Known Activations