INDEX

Explanations

the word "Where" or variations thereof, indicating inquiries about location or position

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

ega

-0.17

adors

-0.16

tings

-0.15

irma

-0.15

å¤ªéĥİ

-0.14

ÑĢÐ°Ð¿

-0.14

sters

-0.14

ifer

-0.14

amm

-0.14

adores

-0.13

POSITIVE LOGITS

abouts

0.17

iyim

0.15

Ð¾Ð¿Ð¸Ñģ

0.14

onis

0.14

bij

0.14

itzer

0.14

 ãĤ°

0.14

istrovstvÃŃ

0.14

 lá»Ńa

0.14

à¸©

0.14

Activations Density 0.020%

the word "Where" or variations thereof, indicating inquiries about location or position

No Comments

No Known Activations