INDEX

Explanations

complex and potentially specific terms related to a particular subject matter, possibly in a technical or academic context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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¿Ð¾Ð±Ð°

-0.24

gress

-0.16

 certo

-0.14

itude

-0.14

 multiline

-0.14

 Ð½Ð°Ð¿ÑĢÐ¸

-0.14

rig

-0.13

 magnitude

-0.13

iverse

-0.13

 correl

-0.13

POSITIVE LOGITS

 Ð¸ÑģÐ¿Ð¾Ð»ÑĮÐ·Ð¾Ð²Ð°Ð½Ð¸

0.21

 ÑįÐºÑģÐ¿Ð»ÑĥÐ°ÑĤÐ°

0.21

 Ð¾ÑĢÐ³Ð°Ð½Ð¸Ð·Ð°

0.18

ÐĲÑĢÑħÑĸÐ²

0.18

ì¦Į

0.16

 ÑįÑĦÑĦÐµÐº

0.16

 à¸ģà¸²à¸£

0.16

 ÑĢÐ°Ð·ÑĢÐ°Ð±Ð¾ÑĤ

0.14

 addCriterion

0.14

ÐµÑĢÑĤÐ¸

0.14

Activations Density 0.088%

complex and potentially specific terms related to a particular subject matter, possibly in a technical or academic context

No Comments

No Known Activations