INDEX

Explanations

references to time periods and specific dates

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

 Victorian

-0.18

-0.17

WW

-0.15

Û²Û°Û±

-0.15

zac

-0.14

Enh

-0.14

forth

-0.14

river

-0.14

olik

-0.14

POSITIVE LOGITS

0.84

0.73

Û±Û¹Û¶

0.54

Û±Û¹Û·

0.49

0.46

 Nixon

0.39

0.36

Û±Û¹Ûµ

0.27

Û±Û¹Û¸

0.25

 Vietnam

0.23

Activations Density 0.098%

references to time periods and specific dates

No Comments

No Known Activations