INDEX

Explanations

HTML or programming-related terms within a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

Ted

-0.17

 Bureau

-0.15

 Pres

-0.15

onna

-0.14

ì¹´ì§Ģëħ¸

-0.14

 staged

-0.14

-0.13

eneric

-0.13

ehr

-0.13

Ted

-0.13

POSITIVE LOGITS

ukkit

0.16

bk

0.15

erce

0.15

INDER

0.15

è¶

0.15

cretion

0.14

aleza

0.14

 Ð»ÑĸÐº

0.14

webkit

0.13

;k

0.13

Activations Density 0.052%

HTML or programming-related terms within a document

No Comments

No Known Activations