INDEX

Explanations

terms related to specialized fields of study or professions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

agg

-0.16

loff

-0.15

lick

-0.15

nop

-0.13

nicas

-0.13

lining

-0.13

ë°©

-0.13

reachable

-0.13

FFFFFFFF

-0.13

lined

-0.13

POSITIVE LOGITS

acades

0.16

ism

0.14

-fontawesome

0.14

ÐºÐ¾Ð¿

0.14

Å¡nÃŃ

0.14

 EXEMPLARY

0.13

 Coupe

0.13

Sharper

0.13

izzard

0.13

eydi

0.12

Activations Density 0.110%

terms related to specialized fields of study or professions

No Comments

No Known Activations