INDEX

Explanations

instances of significant numerical data or metrics

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

↵

-0.23

-0.20

-0.19

 particularly

-0.18

...

-0.17

-t

-0.17

-m

-0.17

 particular

-0.16

POSITIVE LOGITS

yonel

0.20

 There

0.20

istrovstvÃŃ

0.20

ï»¿↵

0.20

Ã¡zev

0.19

 ↵↵

0.19

 This

0.18

It

0.17

Its

0.17

Both

0.16

Activations Density 2.113%

instances of significant numerical data or metrics

No Comments

No Known Activations