INDEX

Explanations

expressions of requests and communication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 supposedly

-0.17

 hence

-0.16

 deemed

-0.16

 albeit

-0.16

orough

-0.15

 allegedly

-0.14

.vo

-0.14

Upon

-0.14

 Hence

-0.14

 Throughout

-0.14

POSITIVE LOGITS

 begin

0.21

 near

0.19

 nearly

0.18

 begins

0.17

bec

0.17

 enjo

0.17

 began

0.16

 comport

0.16

 beginning

0.16

near

0.16

Activations Density 0.079%

expressions of requests and communication

No Comments

No Known Activations