INDEX

Explanations

specific symbols or characters that may represent unique entities or categories

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

_ASCII

-0.13

owell

-0.13

 countless

-0.12

å¢

-0.12

 deÄŁerli

-0.12

 Ð¾Ð³ÑĢÐ¾Ð¼

-0.12

 affair

-0.12

 huge

-0.12

APT

-0.12

ãĥ©ãĥĥãĤ¯

-0.12

POSITIVE LOGITS

 specified

0.27

 statutory

0.26

 stip

0.25

 limitations

0.25

 pursuant

0.25

 limitation

0.25

 periods

0.24

 limits

0.23

 eligibility

0.23

 restrictions

0.23

Activations Density 0.028%

specific symbols or characters that may represent unique entities or categories

No Comments

No Known Activations