INDEX

Explanations

references to London

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

adu

-0.18

backs

-0.16

atever

-0.15

inel

-0.15

fas

-0.14

ui

-0.14

å²¸

-0.14

tings

-0.13

agon

-0.13

POSITIVE LOGITS

ers

0.24

 Bridge

0.21

-based

0.18

istan

0.18

erry

0.18

 Underground

0.18

etta

0.17

ale

0.17

subpackage

0.17

 Metropolitan

0.17

Activations Density 0.013%

references to London

No Comments

No Known Activations