INDEX

Explanations

specific names or titles related to entertainment or media, particularly in a non-English context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

.MixedReality

-0.16

 Tale

-0.14

 LENG

-0.14

.scalablytyped

-0.14

ovich

-0.13

oz

-0.13

inia

-0.13

akeup

-0.13

§è¡Į

-0.13

itus

-0.13

POSITIVE LOGITS

 ÑģÐ²Ð¸Ð´

0.15

Ð½ÑĮÐ¾

0.15

ÃĦ

0.15

illac

0.14

ê³ł

0.14

cion

0.14

rna

0.14

uyÃªn

0.14

 ÐºÐ¾Ð½ÑĤÑĢÐ°

0.14

ê±°ëĤĺ

0.14

Activations Density 0.001%

specific names or titles related to entertainment or media, particularly in a non-English context

No Comments

No Known Activations