INDEX

Explanations

elements related to completion and conclusion 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 @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

-0.16

edor

-0.16

 Welfare

-0.15

Ð¸Ð½Ð¾Ðº

-0.15

 sing

-0.15

 clearance

-0.15

 Sing

-0.14

ountry

-0.14

anne

-0.14

 welfare

-0.14

POSITIVE LOGITS

Occurred

0.17

sWith

0.17

LEGRO

0.16

¶Į

0.16

ë§ī

0.15

inalg

0.15

bah

0.14

ãĤ©

0.14

ãĤ¤ãĥī

0.14

SetActive

0.14

Activations Density 0.126%

elements related to completion and conclusion in various contexts

No Comments

No Known Activations