INDEX

Explanations

references to the term "jur" or related concepts, particularly in legal or contextual frameworks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ness

-0.17

sw

-0.16

chai

-0.16

ings

-0.15

à¸ļà¸²à¸Ĺ

-0.14

uity

-0.14

sms

-0.14

skill

-0.14

ened

-0.14

es

-0.14

POSITIVE LOGITS

assic

0.19

isdiction

0.17

.scalablytyped

0.16

ror

0.15

llib

0.15

antee

0.15

neys

0.15

semble

0.15

isper

0.15

usalem

0.14

Activations Density 0.010%

references to the term "jur" or related concepts, particularly in legal or contextual frameworks

No Comments

No Known Activations