INDEX

Explanations

references to events and activities in Chicago

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

 marked

-0.18

Ø³Ø§ÙĨ

-0.15

AndView

-0.15

avo

-0.15

 Heights

-0.14

 illness

-0.14

rix

-0.14

boro

-0.14

Luz

-0.14

èĥİ

-0.14

POSITIVE LOGITS

 Navy

0.21

 Buckingham

0.19

 Soldier

0.19

 Loop

0.18

 millenn

0.18

 Randolph

0.18

 Near

0.18

Loop

0.17

.hwp

0.17

 Gree

0.17

Activations Density 0.031%

references to events and activities in Chicago

No Comments

No Known Activations