INDEX

Explanations

punctuation and dialogue markers used in conversational contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

Recorder

-0.15

estre

-0.15

itom

-0.15

Appending

-0.14

inos

-0.14

 “âĢ¦

-0.14

ciz

-0.14

lander

-0.13

Î»Î¬

-0.13

à¤¾:

-0.13

POSITIVE LOGITS

 said

0.39

 says

0.35

say

0.28

said

0.27

 Says

0.24

says

0.24

è¯´

0.21

 according

0.21

æĽ°

0.19

 ÑģÐºÐ°Ð·Ð°Ð»

0.19

Activations Density 0.069%

punctuation and dialogue markers used in conversational contexts

No Comments

No Known Activations