INDEX

Explanations

references to horse racing events and participants

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

ÛĮÙĨÙĩ

-0.16

leaflet

-0.15

leaf

-0.15

ocs

-0.15

achel

-0.15

izza

-0.15

izont

-0.14

lige

-0.14

 relic

-0.14

bij

-0.14

POSITIVE LOGITS

 trainer

0.42

 trainers

0.35

 Trainer

0.35

 horses

0.35

trainer

0.33

 rider

0.32

 training

0.30

 stable

0.30

 horse

0.29

 trained

0.28

Activations Density 0.031%

references to horse racing events and participants

No Comments

No Known Activations