INDEX

Explanations

punctuation marks and sentence boundaries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

plus

-0.15

osci

-0.14

å

-0.14

 vice

-0.14

ocre

-0.14

PLUS

-0.13

ifestyles

-0.13

egin

-0.13

æĹ¢çĦ¶

-0.13

Ã«l

-0.13

POSITIVE LOGITS

 behind

0.21

 across

0.21

 minutes

0.20

minutes

0.16

eton

0.16

 inside

0.16

 Minutes

0.15

 ÅĻÃŃj

0.15

 Emerging

0.15

 later

0.15

Activations Density 0.338%

punctuation marks and sentence boundaries

No Comments

No Known Activations