INDEX

Explanations

references to interviews and reports related to health topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ÐºÐ¾Ð·

-0.17

_closure

-0.16

ŀæĢ§

-0.14

ÙĦÙĪ

-0.14

 chamber

-0.14

owie

-0.14

ãĤ¹ãĥĨ

-0.13

 Samuel

-0.13

oga

-0.13

 Scal

-0.13

POSITIVE LOGITS

andler

0.16

tsy

0.15

outu

0.15

tas

0.15

ervo

0.14

ÌĨ

0.14

tog

0.14

ocode

0.14

hani

0.14

obl

0.14

Activations Density 0.008%

references to interviews and reports related to health topics

No Comments

No Known Activations