INDEX

Explanations

terms related to health effects and research studies concerning specific substances or extracts

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

ight

-0.17

aw

-0.17

/read

-0.16

oby

-0.16

-Russian

-0.16

 readily

-0.15

 rigorous

-0.15

Body

-0.15

oy

-0.15

rey

-0.15

POSITIVE LOGITS

efined

0.23

ourke

0.20

cliffe

0.19

aldo

0.18

nant

0.18

asmus

0.18

ansom

0.18

entrant

0.18

neck

0.16

shaw

0.16

Activations Density 2.273%

terms related to health effects and research studies concerning specific substances or extracts

No Comments

No Known Activations