INDEX

Explanations

legal terminology and references to court proceedings

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

 appellate

-0.17

 Appeal

-0.14

 attenu

-0.14

ija

-0.13

 fever

-0.13

 Fellow

-0.13

ogen

-0.13

ughs

-0.13

 Fever

-0.13

inery

-0.13

POSITIVE LOGITS

 federal

0.30

 district

0.25

 Eastern

0.24

district

0.23

US

0.22

istrate

0.21

 District

0.21

 United

0.21

 Federal

0.21

 Northern

0.20

Activations Density 0.039%

legal terminology and references to court proceedings

No Comments

No Known Activations