INDEX

Explanations

phrases describing secrets or keys to success

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orthy

-0.16

 stakes

-0.15

ually

-0.15

las

-0.15

uj

-0.14

WithValue

-0.14

.lp

-0.14

.wordpress

-0.14

 terme

-0.14

LOPT

-0.13

POSITIVE LOGITS

 secret

0.56

 secrets

0.52

 keys

0.44

key

0.43

 Secrets

0.43

 Secret

0.41

Secret

0.41

secret

0.40

 SECRET

0.38

ç§ĺ

0.37

Activations Density 0.115%

phrases describing secrets or keys to success

No Comments

No Known Activations