INDEX

Explanations

references to specific films or projects

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

naments

-0.16

 é©

-0.15

layers

-0.15

allet

-0.15

Ø®ØµÙĪØµ

-0.14

elite

-0.14

áº¯t

-0.14

oria

-0.14

eler

-0.14

vasive

-0.14

POSITIVE LOGITS

til

0.23

 mirror

0.23

 mount

0.21

Til

0.20

 Mount

0.20

 body

0.20

APS

0.19

 Mirror

0.19

 bodies

0.18

mirror

0.18

Activations Density 0.002%

references to specific films or projects

No Comments

No Known Activations