INDEX

Explanations

references to measurement or capacity indicators, particularly in relation to loading or carrying

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

ophon

-0.19

assen

-0.15

asive

-0.15

ropy

-0.14

 Chance

-0.14

 Äĥn

-0.14

ing

-0.14

ode

-0.14

cope

-0.13

ober

-0.13

POSITIVE LOGITS

shed

0.27

-bearing

0.27

ings

0.26

Unload

0.25

bearing

0.21

able

0.20

edException

0.20

edImage

0.20

runner

0.20

 shedding

0.19

Activations Density 0.035%

references to measurement or capacity indicators, particularly in relation to loading or carrying

No Comments

No Known Activations