INDEX

Explanations

instances of the word "dé" and its variations in contexts related to construction or modification

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

ins

-0.15

bane

-0.15

cpy

-0.14

 Lomb

-0.14

Ras

-0.14

Rnd

-0.14

iswa

-0.14

 dostup

-0.14

alchemy

-0.14

al

-0.13

POSITIVE LOGITS

rou

0.21

rog

0.21

concert

0.20

pend

0.19

licate

0.19

limit

0.19

termin

0.18

amb

0.18

lib

0.18

charge

0.18

Activations Density 0.002%

instances of the word "dé" and its variations in contexts related to construction or modification

No Comments

No Known Activations