INDEX

Explanations

names and specified characteristics of people and entities

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

umbo

-0.18

aal

-0.16

ienda

-0.16

lÃŃ

-0.15

ÑĴ

-0.14

IMER

-0.14

 quotation

-0.14

 Milli

-0.14

ãĥ³ãĤº

-0.14

 bord

-0.14

POSITIVE LOGITS

 latter

0.18

 especially

0.17

olon

0.16

especially

0.16

 alone

0.15

Ð½Ð¸Ð½

0.15

 notamment

0.15

 Ø§ÛĮÙĨÚ©Ùĩ

0.15

 particularly

0.15

 Pump

0.14

Activations Density 0.180%

names and specified characteristics of people and entities

No Comments

No Known Activations