INDEX

Explanations

discussions related to creative expression and writing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

Ð¶Ð°

-0.16

itest

-0.15

è¨Ī

-0.15

rita

-0.15

elin

-0.15

hti

-0.14

 Mess

-0.14

 Calc

-0.14

ãĤ¿ãĥ«

-0.14

ines

-0.14

POSITIVE LOGITS

owie

0.17

NGX

0.16

sonian

0.14

dak

0.14

InMillis

0.14

 useClass

0.14

rych

0.14

ãĥ¡ãĥ³ãĥĪ

0.14

nou

0.14

 Nich

0.14

Activations Density 0.302%

discussions related to creative expression and writing

No Comments

No Known Activations