INDEX

Explanations

instances of the word "addressed" and its variations in the context of speeches or statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

resa

-0.17

ãĥĥ

-0.17

eny

-0.17

elim

-0.16

anggan

-0.14

cede

-0.14

isson

-0.14

nila

-0.14

lan

-0.14

fold

-0.14

POSITIVE LOGITS

aten

0.16

ç·Ĭ

0.15

legen

0.15

atorio

0.15

otte

0.14

ator

0.14

oux

0.14

 indeb

0.14

oused

0.14

Activations Density 0.031%

instances of the word "addressed" and its variations in the context of speeches or statements

No Comments

No Known Activations