INDEX

Explanations

names of people, particularly authors or contributors mentioned in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

apons

-0.15

Č

-0.15

Ð¶Ð°

-0.14

yonel

-0.14

åħ·

-0.14

Ã´n

-0.14

ROC

-0.14

agate

-0.14

ÌĨ

-0.14

essen

-0.14

POSITIVE LOGITS

lid

0.15

 abuse

0.14

è¡¥

0.14

adaki

0.13

oulos

0.13

hum

0.13

orden

0.12

à¸´à¸Ļà¸Ĺà¸£

0.12

Boh

0.12

oken

0.12

Activations Density 0.065%

names of people, particularly authors or contributors mentioned in the text

No Comments

No Known Activations