INDEX

Explanations

legal or procedural terminology related to court rulings and their implications

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

-0.17

ÑĢÐ°Ñħ

-0.15

alo

-0.14

å¢

-0.14

 Trot

-0.14

ica

-0.14

 sadd

-0.13

 Ø´ÙĪ

-0.13

deg

-0.13

utex

-0.13

POSITIVE LOGITS

 invalid

0.45

 challenge

0.41

invalid

0.37

 null

0.37

 invalidate

0.36

 Invalid

0.36

 challenged

0.35

 void

0.35

 Challenge

0.35

 INVALID

0.33

Activations Density 0.130%

legal or procedural terminology related to court rulings and their implications

No Comments

No Known Activations