INDEX

Explanations

punctuation marks 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 @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

hend

-0.16

zet

-0.16

ossier

-0.16

ormsg

-0.15

 ragaz

-0.15

.updateDynamic

-0.14

¶ģ

-0.14

icolon

-0.14

#ga

-0.14

$MESS

-0.14

POSITIVE LOGITS

 there

0.20

eut

0.17

lez

0.16

 respectively

0.15

there

0.15

se

0.15

we

0.15

ides

0.13

etc

0.13

 Vert

0.13

Activations Density 0.202%

punctuation marks in the text

No Comments

No Known Activations