INDEX

Explanations

special characters or unusual symbols used in the text

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

awks

-0.16

 ilan

-0.16

ilan

-0.15

rowse

-0.15

ocache

-0.15

oad

-0.14

ilder

-0.14

il

-0.14

Ð¾ÑĢÐ¾ÑĤ

-0.14

 Dise

-0.13

POSITIVE LOGITS

 people

0.17

SW

0.15

 southwest

0.15

people

0.15

ngine

0.15

Syn

0.15

 under

0.14

 People

0.14

 Under

0.14

under

0.14

Activations Density 0.010%

special characters or unusual symbols used in the text

No Comments

No Known Activations