INDEX

Explanations

terms related to caching and cache management

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

loid

-0.15

uria

-0.14

itÃ¤

-0.14

olle

-0.14

oll

-0.14

ieg

-0.14

ylim

-0.14

_Anim

-0.14

allis

-0.13

ollapse

-0.13

POSITIVE LOGITS

buster

0.28

able

0.27

ingly

0.24

busters

0.22

ABLE

0.20

-Control

0.19

ability

0.18

ables

0.17

abl

0.17

åĻ¨

0.16

Activations Density 0.026%

terms related to caching and cache management

No Comments

No Known Activations