INDEX

Explanations

references to locations and institutions in Arizona

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oda

-0.19

 Ches

-0.17

atta

-0.16

/packages

-0.15

onde

-0.15

RTE

-0.14

 belg

-0.14

 BITS

-0.14

.microsoft

-0.14

 Belg

-0.14

POSITIVE LOGITS

 Arizona

0.19

 desert

0.17

Nev

0.16

Arizona

0.16

 Phoenix

0.16

 Nevada

0.16

itest

0.15

 Desert

0.15

unas

0.15

áºŃu

0.15

Activations Density 0.404%

references to locations and institutions in Arizona

No Comments

No Known Activations