INDEX

Explanations

references to the fictional nature of the work and disclaimers regarding real-life resemblances

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

avor

-0.18

 Proud

-0.15

eman

-0.15

illin

-0.15

men

-0.15

aux

-0.14

 Arts

-0.14

ï

-0.14

bond

-0.14

(Paint

-0.14

POSITIVE LOGITS

letes

0.16

.appspot

0.16

 Schmidt

0.16

yme

0.15

utra

0.15

mÃŃt

0.15

EK

0.14

.DialogResult

0.14

edm

0.14

lette

0.14

Activations Density 0.008%

references to the fictional nature of the work and disclaimers regarding real-life resemblances

No Comments

No Known Activations

references to the fictional nature of the work and disclaimers regarding real-life resemblances

No Comments

No Known Activations