INDEX

Explanations

specific names and identifiers, likely in the context of citations or bibliographies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

 Word

-0.15

ãĥ³ãĥĢ

-0.14

 aver

-0.14

.bitmap

-0.14

ext

-0.14

veis

-0.14

 Hunts

-0.14

unk

-0.14

 Wort

-0.14

POSITIVE LOGITS

 Forget

0.23

 Bureau

0.22

 Martial

0.21

 Rena

0.20

 Legend

0.20

 Colin

0.19

Dup

0.19

 Rica

0.19

DUP

0.19

 Blonde

0.19

Activations Density 0.100%

specific names and identifiers, likely in the context of citations or bibliographies

No Comments

No Known Activations