INDEX

Explanations

concepts related to comparison and pairing in various contexts

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

Äįan

-0.17

 DÃ¼n

-0.16

ÑĢÐ¾Ð²Ð¸Ñĩ

-0.15

inker

-0.14

Toe

-0.14

esel

-0.14

zew

-0.13

alah

-0.13

 thumb

-0.13

/run

-0.13

POSITIVE LOGITS

illard

0.17

asonry

0.15

 giá»¯a

0.14

ernet

0.14

abbo

0.14

ÙĪØ·

0.14

reno

0.14

0.13

abella

0.13

cou

0.13

Activations Density 0.286%

concepts related to comparison and pairing in various contexts

No Comments

No Known Activations