INDEX

Explanations

phrases related to historical artifacts and their provenance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

/fw

-0.15

endir

-0.14

marsh

-0.14

Lad

-0.14

aggi

-0.14

-sector

-0.14

 direction

-0.13

pent

-0.13

.worker

-0.13

Luk

-0.13

POSITIVE LOGITS

eneg

0.16

koc

0.15

éĳĳ

0.14

eyse

0.14

CLUSIVE

0.14

ConfigurationException

0.14

 mastur

0.14

çīĮ

0.13

ãĤº

0.13

 INCIDENT

0.13

Activations Density 0.450%

phrases related to historical artifacts and their provenance

No Comments

No Known Activations