INDEX

Explanations

references to future time periods and durations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

Ø·Ø©

-0.15

okit

-0.14

angl

-0.14

ordin

-0.14

readcr

-0.14

legt

-0.13

yor

-0.13

nech

-0.13

ÑĢÐ¾ÑģÑĤÐ¾

-0.13

.ua

-0.13

POSITIVE LOGITS

 ahead

0.33

 come

0.31

yet

0.30

ahead

0.25

 Ahead

0.24

 follow

0.24

 hence

0.24

 unborn

0.24

 decades

0.24

 years

0.23

Activations Density 0.030%

references to future time periods and durations

No Comments

No Known Activations