INDEX

Explanations

words with a focus on proper nouns related to names and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

oftware

-0.19

ingham

-0.19

azÄĥ

-0.16

abyrinth

-0.16

oft

-0.16

Å¡ov

-0.15

ahoma

-0.15

Ĺi

-0.14

BUTTONDOWN

-0.14

¿

-0.14

POSITIVE LOGITS

ht

0.16

isk

0.15

FF

0.15

-fw

0.14

IFF

0.14

 Bruno

0.14

elt

0.14

htar

0.13

adal

0.13

HT

0.13

Activations Density 0.084%

words with a focus on proper nouns related to names and titles

No Comments

No Known Activations