INDEX

Explanations

references to music releases and record labels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

oder

-0.16

å¼ı

-0.15

ordion

-0.14

iyon

-0.14

abi

-0.14

ufen

-0.14

lander

-0.14

igraph

-0.13

obus

-0.13

abis

-0.13

POSITIVE LOGITS

 Island

0.25

 Columbia

0.25

 Warner

0.25

 Poly

0.25

 Universal

0.24

 Atlantic

0.24

 Koch

0.24

Atlantic

0.23

BM

0.23

BM

0.23

Activations Density 0.051%

references to music releases and record labels

No Comments

No Known Activations