INDEX

Explanations

features and amenities of a property

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ix

-0.16

 Australians

-0.16

 ontvangst

-0.15

 condos

-0.15

åħ¥ãĤĮ

-0.15

 cust

-0.15

ric

-0.15

 Stick

-0.14

å¿½

-0.14

ANGES

-0.14

POSITIVE LOGITS

 undercover

0.21

 entertain

0.19

 entert

0.19

 Internal

0.17

 internal

0.17

alf

0.17

Internal

0.17

ens

0.16

 timber

0.16

 reverse

0.16

Activations Density 0.037%

features and amenities of a property

No Comments

No Known Activations