INDEX

Explanations

references to architectural professionals and their credentials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

etto

-0.18

agem

-0.18

å¼¥

-0.16

egin

-0.15

Ð´Ð°Ð¼

-0.15

onse

-0.14

Ð¿ÑĸÐ¾Ð½

-0.14

isons

-0.14

implify

-0.14

idad

-0.14

POSITIVE LOGITS

ARB

0.18

 Architect

0.16

 architecture

0.15

 Perkins

0.15

 preservation

0.15

 Store

0.15

 mies

0.15

 Preservation

0.15

 starch

0.15

ARCH

0.14

Activations Density 0.054%

references to architectural professionals and their credentials

No Comments

No Known Activations