INDEX

Explanations

time-related information presented in a specific format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

nap

-0.17

ulan

-0.17

itle

-0.16

Ã©mon

-0.15

 COVID

-0.15

 mechanism

-0.14

 Jacob

-0.13

adel

-0.13

omik

-0.13

POSITIVE LOGITS

áºł

0.17

kara

0.17

aksi

0.15

ighest

0.15

.SDK

0.15

 Bilim

0.14

luluk

0.14

uml

0.14

tridge

0.14

ajs

0.14

Activations Density 0.041%

time-related information presented in a specific format

No Comments

No Known Activations

time-related information presented in a specific format

No Comments

No Known Activations