INDEX

Explanations

specific numeric or highly technical references within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 behaviors

-0.18

neighbor

-0.18

 neighborhood

-0.17

 rumored

-0.17

favor

-0.16

oward

-0.16

 neighborhoods

-0.16

 Neighborhood

-0.16

 neighbors

-0.16

armor

-0.16

POSITIVE LOGITS

 Malays

0.27

 Malaysian

0.25

 Malaysia

0.22

 Malay

0.22

UM

0.20

 Ministers

0.19

 Chinese

0.19

 Kuala

0.18

Naj

0.17

Pak

0.17

Activations Density 0.002%

specific numeric or highly technical references within the text

No Comments

No Known Activations