INDEX

Explanations

elements related to storytelling and narrative structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

usage

-0.15

ãĤĪãģĨ

-0.14

zzo

-0.14

ahlen

-0.14

fol

-0.14

azon

-0.13

succ

-0.13

aton

-0.13

ayan

-0.13

utzer

-0.13

POSITIVE LOGITS

 stories

0.24

 tales

0.20

 story

0.20

stories

0.18

 accounts

0.18

 history

0.18

 Stories

0.18

æķħäºĭ

0.17

 tale

0.17

Stories

0.17

Activations Density 0.326%

elements related to storytelling and narrative structure

No Comments

No Known Activations