INDEX

Explanations

words associated with technical terminology and science-related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

enberg

-0.18

abol

-0.16

 prime

-0.15

adin

-0.15

ardy

-0.15

argas

-0.14

 Prime

-0.14

 elev

-0.14

 Love

-0.14

abolic

-0.14

POSITIVE LOGITS

vale

0.16

outed

0.15

mania

0.14

.shtml

0.14

-Version

0.14

 Mixed

0.14

chied

0.14

ScreenState

0.14

mixed

0.14

vib

0.14

Activations Density 0.002%

words associated with technical terminology and science-related concepts

No Comments

No Known Activations