INDEX

Explanations

sequences of numbers or terms that denote order or classification

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

ication

-0.15

íĸ¥

-0.15

-0.14

ÐµÐ»Ñİ

-0.14

erdem

-0.14

ritz

-0.14

Ð°Ð¹ÑĤ

-0.14

essler

-0.14

land

-0.13

POSITIVE LOGITS

ancode

0.16

Diagram

0.16

ighb

0.16

DAG

0.15

amera

0.15

dana

0.15

undi

0.14

encer

0.14

OF

0.14

ively

0.14

Activations Density 0.019%

sequences of numbers or terms that denote order or classification

No Comments

No Known Activations