INDEX

Explanations

terms related to monitoring and tracking performance or progress in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ios

-0.16

yi

-0.15

marvin

-0.15

anik

-0.14

ulos

-0.14

omba

-0.14

kÃ³w

-0.14

edback

-0.14

IOS

-0.14

_locked

-0.13

POSITIVE LOGITS

 progress

0.26

 closely

0.25

 track

0.20

 Progress

0.19

-progress

0.19

track

0.18

 trends

0.18

 Track

0.18

-track

0.18

progress

0.18

Activations Density 0.066%

terms related to monitoring and tracking performance or progress in various contexts

No Comments

No Known Activations