INDEX

Explanations

interactions involving greetings and friendly exchanges

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

phem

-0.17

uar

-0.16

bab

-0.15

.Framework

-0.15

ucas

-0.15

 xlink

-0.14

onas

-0.14

èĮĤ

-0.14

oran

-0.14

gent

-0.14

POSITIVE LOGITS

0.17

 hello

0.15

å¯Ĵ

0.15

 greet

0.15

ozÃŃ

0.15

endir

0.14

 Neck

0.14

Dix

0.14

 greetings

0.13

Activations Density 0.199%

interactions involving greetings and friendly exchanges

No Comments

No Known Activations