INDEX

Explanations

effective strategies and methods used in therapy and teaching

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Wich

-0.16

QualifiedName

-0.15

ä¸ĺ

-0.15

Ð±ÑĥÐ´ÑĮ

-0.14

etrics

-0.14

udden

-0.14

udies

-0.14

RIX

-0.14

edn

-0.14

izik

-0.14

POSITIVE LOGITS

uman

0.14

å»·

0.13

 Expression

0.13

 contr

0.13

omet

0.13

obec

0.13

oire

0.13

current

0.12

UMAN

0.12

à¤«

0.12

Activations Density 0.531%

effective strategies and methods used in therapy and teaching

No Comments

No Known Activations