INDEX

Explanations

references to academic or professional fellowships and awards

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

rok

-0.17

Ð¾ÑĩÐ½Ð¾

-0.15

anches

-0.15

 ActionTypes

-0.15

anyak

-0.14

CTest

-0.14

Sequential

-0.14

amac

-0.14

alternate

-0.14

olicit

-0.14

POSITIVE LOGITS

ships

0.29

ship

0.20

exion

0.18

SHIP

0.17

hood

0.16

shipping

0.15

ÙĨØªÛĮ

0.15

ÑĢÑĥÐ½

0.15

ynes

0.14

Activations Density 0.007%

references to academic or professional fellowships and awards

No Comments

No Known Activations