INDEX

Explanations

references to characters and their roles within a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

owski

-0.17

arent

-0.15

icontrol

-0.15

afort

-0.15

å®

-0.15

å»

-0.14

ìĿ´íĦ°

-0.14

roti

-0.14

kus

-0.14

 ancient

-0.13

POSITIVE LOGITS

Ing

0.21

Ing

0.17

 Alman

0.17

ç·

0.16

adia

0.15

 Torrent

0.15

phans

0.15

ippi

0.15

 Baum

0.15

hom

0.14

Activations Density 0.027%

references to characters and their roles within a narrative context

No Comments

No Known Activations