INDEX

Explanations

terms related to judging and evaluating in a competitive context

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

ixels

-0.14

atcher

-0.14

imiz

-0.13

paren

-0.13

elage

-0.13

ãĥ¼ãĥĭ

-0.13

ÅĽmy

-0.13

çŃĴ

-0.13

Ã¼re

-0.13

MediaPlayer

-0.12

POSITIVE LOGITS

 ëĭ¤ìļ´ë°Ľê¸°

0.15

 fucking

0.14

 ...↵

0.14

 cock

0.14

cil

0.14

 âĢ¦↵

0.14

0.13

ogui

0.13

ÂŃi

0.13

ãĢħ

0.13

Activations Density 11.140%

terms related to judging and evaluating in a competitive context

No Comments

No Known Activations