INDEX

Explanations

technical terms related to configuration and network settings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

_DOM

-0.16

 pilgrimage

-0.15

agna

-0.14

ekim

-0.14

 archit

-0.14

rdf

-0.14

.Pixel

-0.14

HTTP

-0.14

 Hartford

-0.13

fits

-0.13

POSITIVE LOGITS

 bond

0.25

 Bond

0.21

 Metric

0.21

eth

0.21

 metric

0.20

bond

0.20

 ether

0.20

 link

0.20

 Link

0.19

 scope

0.19

Activations Density 0.006%

technical terms related to configuration and network settings

No Comments

No Known Activations