INDEX

Explanations

references to radio shows and broadcasts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

-INF

-0.16

locs

-0.14

 USPS

-0.14

ÏĥÎ¼Î±

-0.14

olen

-0.14

 ÑĤÐµÐ»

-0.14

agus

-0.14

 rÃ³wn

-0.13

 Pony

-0.13

/goto

-0.13

POSITIVE LOGITS

 listeners

0.27

 station

0.26

 stations

0.26

 morning

0.25

 afternoon

0.24

 mornings

0.24

 listener

0.23

 radio

0.23

0.22

 Arbit

0.22

Activations Density 0.036%

references to radio shows and broadcasts

No Comments

No Known Activations