INDEX

Explanations

references to governmental positions and relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ackers

-0.19

ibar

-0.15

-upper

-0.15

.tell

-0.15

è¤

-0.15

ipt

-0.14

¤í

-0.14

ÙĤÙĩ

-0.14

á»¥

-0.14

 diplom

-0.14

POSITIVE LOGITS

vic

0.16

isch

0.15

VD

0.14

ow

0.14

odi

0.14

onet

0.14

tog

0.14

ÎµÎ¹

0.14

brook

0.14

ucher

0.14

Activations Density 0.406%

references to governmental positions and relationships

No Comments

No Known Activations