INDEX

Explanations

references to online archival or historical data sources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

usat

-0.16

 hala

-0.16

lsen

-0.15

á»¡

-0.15

EventData

-0.15

:↵↵↵↵↵↵

-0.14

 Bauer

-0.14

erde

-0.14

igor

-0.14

artner

-0.14

POSITIVE LOGITS

0.16

æĮº

0.16

 extr

0.15

.scalablytyped

0.15

Slinky

0.14

Ð»ÐµÑĤ

0.14

ules

0.14

RLF

0.14

cale

0.14

Activations Density 0.067%

references to online archival or historical data sources

No Comments

No Known Activations