INDEX

Explanations

negations or limitations 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 @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

 sharp

-0.19

leigh

-0.17

 youngest

-0.16

sharp

-0.16

 weakest

-0.15

 Sharp

-0.14

 smallest

-0.14

Sharp

-0.14

 fairly

-0.14

POSITIVE LOGITS

 more

0.60

æĽ´å¤ļ

0.52

 MORE

0.49

more

0.47

 greater

0.47

 lebih

0.46

 Ð±Ð¾Ð»ÑĮÑĪÐµ

0.44

-more

0.42

 daha

0.42

 wiÄĻcej

0.42

Activations Density 0.003%

negations or limitations in the text

No Comments

No Known Activations