INDEX

Explanations

references to time and its usage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ã²

-0.15

ostel

-0.14

 Gross

-0.14

Bias

-0.14

olum

-0.14

á»Ļ

-0.14

Ã³st

-0.14

 chir

-0.13

 Mapper

-0.13

umo

-0.13

POSITIVE LOGITS

 boring

0.17

 boredom

0.17

activex

0.17

 bored

0.16

avage

0.16

 filler

0.15

idis

0.15

éĸ

0.14

USES

0.14

Activations Density 0.109%

references to time and its usage

No Comments

No Known Activations