INDEX

Explanations

dates and numerical values

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

alam

-0.17

åįł

-0.15

ikan

-0.15

ollapsed

-0.14

iali

-0.14

 libert

-0.14

idian

-0.14

 Prom

-0.14

ubat

-0.14

-eyed

-0.13

POSITIVE LOGITS

chas

0.18

inous

0.16

 extrad

0.15

 topl

0.15

 Codable

0.15

stell

0.14

Ð¿Ð¾Ð²

0.14

Īëĭ¤

0.14

SIG

0.13

anh

0.13

Activations Density 0.047%

dates and numerical values

No Comments

No Known Activations