INDEX

Explanations

technical jargon or references related to software errors and configurations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 çĹ

-0.15

ina

-0.15

pole

-0.14

.au

-0.14

 lodged

-0.14

 ferment

-0.14

 Fare

-0.14

 ãĥĬ

-0.14

Ø±ÛĮØ²

-0.13

agal

-0.13

POSITIVE LOGITS

eth

0.30

 interface

0.27

 bond

0.27

 Interface

0.26

ETH

0.26

 interfaces

0.25

Eth

0.24

 Interfaces

0.24

Eth

0.23

 bonding

0.23

Activations Density 0.013%

technical jargon or references related to software errors and configurations

No Comments

No Known Activations