INDEX

Explanations

instances where something occurs or is executed in the context of an action or a state change

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ropri

-0.16

eyle

-0.16

osti

-0.16

uco

-0.16

latlong

-0.15

 kÄ±

-0.15

.MixedReality

-0.15

gradable

-0.15

adm

-0.14

ofday

-0.14

POSITIVE LOGITS

 demand

0.26

 startup

0.25

_demand

0.22

demand

0.22

fly

0.21

startup

0.20

fly

0.19

-fly

0.19

top

0.19

 runtime

0.18

Activations Density 0.091%

instances where something occurs or is executed in the context of an action or a state change

No Comments

No Known Activations