INDEX

Explanations

terms and phrases related to legal or formal procedural contexts

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

Ã¸re

-0.18

 Frozen

-0.15

asant

-0.15

ooled

-0.14

jab

-0.14

olume

-0.14

æľ«

-0.13

agra

-0.13

/am

-0.13

POSITIVE LOGITS

ãĢħ

0.16

 Protective

0.15

 protective

0.15

ãĥĨãĥ«

0.14

shield

0.14

_READONLY

0.14

 protector

0.14

 Clearance

0.14

 Ð·Ð°ÑīÐ¸ÑĤ

0.14

holders

0.13

Activations Density 0.015%

terms and phrases related to legal or formal procedural contexts

No Comments

No Known Activations