INDEX

Explanations

numerical and time-related references within the text

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

bose

-0.14

Tier

-0.14

uctive

-0.14

horn

-0.14

_migration

-0.14

ottenham

-0.13

udeau

-0.13

Î½Î·

-0.13

egas

-0.13

POSITIVE LOGITS

ervas

0.16

illus

0.16

set

0.15

PPP

0.14

WARDED

0.14

loating

0.14

rian

0.14

 body

0.14

onica

0.14

Jab

0.14

Activations Density 0.001%

numerical and time-related references within the text

No Comments

No Known Activations