INDEX

Explanations

terms related to performance and effectiveness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

hton

-0.18

Î»Î¹Îº

-0.17

Ð»Ð¸ÑĪÐºÐ¾Ð¼

-0.16

alem

-0.16

ssi

-0.15

 Kurul

-0.14

EventArgs

-0.14

ÑĤÐµÐ½

-0.14

oud

-0.14

ophon

-0.14

POSITIVE LOGITS

dur

0.18

 charge

0.16

ez

0.16

 Charge

0.15

ektor

0.15

-performance

0.15

qus

0.15

Dur

0.15

ITA

0.15

LL

0.14

Activations Density 0.083%

terms related to performance and effectiveness

No Comments

No Known Activations