INDEX

Explanations

mentions of "New York" and its variants in various contexts

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

ulus

-0.17

gether

-0.16

itel

-0.16

atur

-0.16

era

-0.16

exus

-0.16

ero

-0.15

ature

-0.14

 poor

-0.14

ucci

-0.14

POSITIVE LOGITS

sik

0.21

shire

0.16

flater

0.16

 City

0.15

:async

0.15

imizer

0.15

ãĥ«ãĥķ

0.14

/New

0.14

æ¬ł

0.14

anness

0.14

Activations Density 0.030%

mentions of "New York" and its variants in various contexts

No Comments

No Known Activations