INDEX

Explanations

punctuation and quotation marks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

-0.14

 lyon

-0.14

ernes

-0.14

-Smith

-0.13

-Israel

-0.13

cpt

-0.13

acre

-0.12

amps

-0.12

 cigaret

-0.12

POSITIVE LOGITS

kus

0.24

kad

0.22

kur

0.22

kas

0.21

Nama

0.21

pak

0.21

kola

0.21

mamak

0.21

maal

0.21

kal

0.21

Activations Density 0.197%

punctuation and quotation marks

No Comments

No Known Activations