INDEX

Explanations

terms related to legal arrests and law enforcement actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 ÐĴÐ¸ÐºÐ¸

-0.16

Ãºng

-0.15

ismatic

-0.15

onian

-0.14

tractor

-0.14

Î³ÎµÎ¹

-0.14

signature

-0.13

ALLY

-0.13

ritel

-0.13

apia

-0.13

POSITIVE LOGITS

 Walsh

0.16

ros

0.16

YLE

0.15

inks

0.14

mage

0.14

 Singleton

0.14

oders

0.14

cona

0.14

 Span

0.14

ynch

0.14

Activations Density 0.000%

terms related to legal arrests and law enforcement actions

No Comments

No Known Activations