INDEX

Explanations

phrases that indicate assistance or support 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 @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 Ð¢Ð°ÐºÐ¸Ð¼

-0.14

ixa

-0.13

 Completely

-0.13

 purely

-0.13

 Alternatively

-0.13

 Other

-0.13

 Miscellaneous

-0.12

 Different

-0.12

 wholly

-0.12

ibri

-0.12

POSITIVE LOGITS

 especially

0.95

 particularly

0.86

especially

0.82

 especialmente

0.81

particularly

0.73

å°¤

0.71

 Ð¾ÑģÐ¾Ð±ÐµÐ½Ð½Ð¾

0.67

çī¹åĪ«

0.67

pecially

0.67

 particular

0.67

Activations Density 0.727%

phrases that indicate assistance or support in various contexts

No Comments

No Known Activations