INDEX

Explanations

terms related to legal services and guidelines

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

stag

-0.17

/browse

-0.17

styl

-0.15

ampp

-0.15

STALL

-0.15

uro

-0.14

agenta

-0.14

 YÃĸ

-0.14

 Conserv

-0.13

 interp

-0.13

POSITIVE LOGITS

 counsel

0.23

 Firm

0.22

 Chambers

0.19

 lawyers

0.18

Lit

0.18

 practice

0.18

 coun

0.17

Debe

0.17

 counseling

0.17

 firm

0.17

Activations Density 0.039%

terms related to legal services and guidelines

No Comments

No Known Activations