INDEX

Explanations

buildings and their historical transformations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

åī¯

-0.17

igest

-0.15

 Fetish

-0.15

arring

-0.15

ÙĪØ±ÙĪØ¨

-0.14

arget

-0.14

 Ð»ÑĸÑĤ

-0.14

asse

-0.14

oidal

-0.14

atform

-0.13

POSITIVE LOGITS

 usage

0.18

usage

0.15

 decades

0.15

decade

0.15

 Lind

0.15

ogn

0.14

 Modern

0.14

igner

0.14

indi

0.14

lage

0.14

Activations Density 0.063%

buildings and their historical transformations

No Comments

No Known Activations