INDEX

Explanations

details related to time durations or scheduling

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Hubbard

-0.16

 thousands

-0.14

chner

-0.14

odel

-0.14

 Eleven

-0.13

 Wochen

-0.13

-years

-0.13

äºĶæľĪ

-0.13

ictim

-0.13

 Months

-0.13

POSITIVE LOGITS

 hour

0.55

 Hour

0.41

-hour

0.41

hour

0.41

 HOUR

0.39

Hour

0.38

 hours

0.37

hr

0.36

.hour

0.35

_hour

0.34

Activations Density 0.088%

details related to time durations or scheduling

No Comments

No Known Activations