INDEX

Explanations

words related to governmental and political entities or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

ä¸»ä»»

-0.16

 sede

-0.16

é³

-0.15

ewise

-0.15

ini

-0.15

icha

-0.15

ppard

-0.14

prung

-0.14

odel

-0.14

ey

-0.14

POSITIVE LOGITS

hood

0.26

-owned

0.23

-Owned

0.22

craft

0.21

-run

0.21

-sponsored

0.20

 Owned

0.20

 owned

0.20

owned

0.19

 sponsored

0.19

Activations Density 0.033%

words related to governmental and political entities or concepts

No Comments

No Known Activations