INDEX

Explanations

phrases and concepts that challenge the novelty of existing ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

 Cool

-0.16

ÏĥÏĩ

-0.16

ekt

-0.15

dbh

-0.14

ected

-0.14

nov

-0.14

 supply

-0.14

upid

-0.13

ended

-0.13

Äı

-0.13

POSITIVE LOGITS

 already

0.20

anel

0.20

already

0.19

ANEL

0.19

Already

0.18

 Already

0.17

 Ð·Ð½Ð°

0.16

 sáºµn

0.15

 schon

0.15

 juÅ¼

0.14

Activations Density 0.207%

phrases and concepts that challenge the novelty of existing ideas

No Comments

No Known Activations