INDEX

Explanations

expressions and phrases indicating perceptions of simplicity or complexity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ÑĤÐµÐ½

-0.16

ughter

-0.14

kus

-0.14

 incompet

-0.14

WARE

-0.14

 incompetent

-0.14

iland

-0.14

 incompetence

-0.14

ehler

-0.14

oad

-0.14

POSITIVE LOGITS

 corn

0.27

 counter

0.22

 presum

0.20

 tack

0.20

 contr

0.20

corn

0.19

 obvious

0.19

counter

0.19

 radical

0.18

odd

0.18

Activations Density 0.190%

expressions and phrases indicating perceptions of simplicity or complexity

No Comments

No Known Activations