INDEX

Explanations

instances of the letter "T" or variations thereof

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

agner

-0.17

Î´Î¯

-0.17

ocus

-0.17

rie

-0.15

ADED

-0.15

ensor

-0.15

izon

-0.15

uple

-0.15

Ã¸re

-0.15

itel

-0.14

POSITIVE LOGITS

ames

0.21

ynes

0.20

ynec

0.18

regon

0.18

regar

0.18

elf

0.17

idy

0.16

ams

0.16

ulse

0.16

atty

0.16

Activations Density 0.014%

instances of the letter "T" or variations thereof

No Comments

No Known Activations

instances of the letter "T" or variations thereof

No Comments

No Known Activations