INDEX

Explanations

concepts related to change and transformation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ÑĮÐµ

-0.14

posables

-0.14

gota

-0.14

anske

-0.13

rup

-0.13

eniable

-0.13

oku

-0.13

.dtd

-0.13

ront

-0.13

anio

-0.12

POSITIVE LOGITS

how

0.90

how

0.75

way

0.68

å¦Ĥä½ķ

0.66

 cÃ³mo

0.61

HOW

0.60

How

0.57

-how

0.54

 nasÄ±l

0.54

How

0.53

Activations Density 0.296%

concepts related to change and transformation

No Comments

No Known Activations