INDEX

Explanations

themes related to conflict and resolution

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 aforementioned

-0.17

exampleInputEmail

-0.14

ÐµÑĢ

-0.13

Ñĵ

-0.13

igs

-0.12

/as

-0.12

æķ£

-0.12

ster

-0.12

Ð¾Ñı

-0.12

POSITIVE LOGITS

sert

0.15

£½

0.15

Msp

0.14

-:-

0.14

ahrenheit

0.14

ramework

0.14

yaw

0.14

AFX

0.14

ibri

0.13

ienes

0.13

Activations Density 1.842%

themes related to conflict and resolution

No Comments

No Known Activations