INDEX

Explanations

phrases indicating a high level of quality or value

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

pras

-0.17

 hence

-0.16

 grand

-0.14

 Hence

-0.14

mani

-0.14

Ð¾Ð¼ÐµÐ½

-0.14

 really

-0.14

ru

-0.14

alon

-0.14

æµ©

-0.13

POSITIVE LOGITS

eur

0.17

acket

0.16

avin

0.15

-mini

0.14

CAF

0.14

uario

0.13

ritz

0.13

hetto

0.13

LS

0.13

izont

0.13

Activations Density 0.016%

phrases indicating a high level of quality or value

No Comments

No Known Activations