INDEX

Explanations

technical terms and phrases related to engineering and diagnostics

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

 famously

-0.19

 challenge

-0.15

impse

-0.15

ikon

-0.15

 leveraging

-0.15

 challenges

-0.15

 Challenges

-0.15

 targeting

-0.15

captures

-0.15

 tweak

-0.15

POSITIVE LOGITS

.Experimental

0.16

č

0.16

emploi

0.16

č

0.15

.degree

0.15

 kinetic

0.15

 experimental

0.15

viz

0.15

osc

0.15

 Banc

0.14

Activations Density 0.020%

technical terms and phrases related to engineering and diagnostics

No Comments

No Known Activations