INDEX

Explanations

instances of high-frequency phrases or elements within text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

onne

-0.15

ather

-0.14

oly

-0.14

uning

-0.14

¢

-0.14

elta

-0.14

McL

-0.14

peria

-0.14

grep

-0.14

hatt

-0.14

POSITIVE LOGITS

Vad

0.16

ogram

0.15

rix

0.14

FAG

0.14

 tslib

0.14

roph

0.14

RPM

0.14

.ix

0.14

nish

0.14

Activations Density 0.004%

instances of high-frequency phrases or elements within text

No Comments

No Known Activations