INDEX

Explanations

references to notable franchises and their extended lore

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

odium

-0.15

ãĤĥ

-0.14

Uploaded

-0.14

/hr

-0.14

hir

-0.14

Scene

-0.13

/std

-0.13

enge

-0.13

nek

-0.13

etch

-0.13

POSITIVE LOGITS

 franchise

0.33

 canon

0.32

 franch

0.29

 property

0.29

 properties

0.28

 universe

0.28

 franchises

0.28

IP

0.27

 univers

0.27

property

0.26

Activations Density 0.161%

references to notable franchises and their extended lore

No Comments

No Known Activations