INDEX

Explanations

temporal references to years or ages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Ukra

-0.16

oldt

-0.15

Ctrls

-0.14

igger

-0.14

ÅĻiv

-0.14

 year

-0.14

laus

-0.14

 period

-0.14

iggers

-0.14

LES

-0.14

POSITIVE LOGITS

 siguiente

0.23

 suiv

0.21

 siguientes

0.20

 preced

0.20

 poster

0.19

achten

0.18

 anterior

0.18

Poster

0.17

poster

0.17

 success

0.17

Activations Density 0.014%

temporal references to years or ages

No Comments

No Known Activations