INDEX

Explanations

phrases that express critique or feedback about experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

atta

-0.15

enstein

-0.14

ãĥ³ãĤ¹

-0.14

aret

-0.14

iska

-0.14

apt

-0.14

anim

-0.14

ÏģÎºÎµ

-0.14

 impress

-0.14

_Master

-0.14

POSITIVE LOGITS

 hÆ¡i

0.17

some

0.15

 slightly

0.15

alue

0.15

agle

0.15

velle

0.15

arkan

0.14

Ø¬Ø§ÙĨ

0.14

 some

0.14

ehler

0.14

Activations Density 0.212%

phrases that express critique or feedback about experiences

No Comments

No Known Activations