INDEX

Explanations

words relating to freshness or recent occurrences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

zon

-0.15

ikal

-0.15

etin

-0.14

reach

-0.14

-notch

-0.14

çĲ³

-0.13

enef

-0.13

 eventual

-0.13

ÐµÐ»Ð¸Ñĩ

-0.13

eward

-0.13

POSITIVE LOGITS

 mint

0.35

mint

0.26

wed

0.26

 Mint

0.20

 formed

0.19

 created

0.19

-formed

0.18

 arrived

0.18

wed

0.16

-created

0.16

Activations Density 0.010%

words relating to freshness or recent occurrences

No Comments

No Known Activations