INDEX

Explanations

time-related information or duration indicators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

etten

-0.17

ston

-0.16

itoris

-0.15

stad

-0.14

à¥Įà¤²

-0.14

ohan

-0.14

missible

-0.14

ILLISECONDS

-0.14

abet

-0.13

Duy

-0.13

POSITIVE LOGITS

ago

0.24

ago

0.20

utter

0.18

uti

0.17

ssf

0.17

(s

0.17

 walk

0.16

abble

0.16

 Thomson

0.16

utters

0.16

Activations Density 0.011%

time-related information or duration indicators

No Comments

No Known Activations