INDEX

Explanations

instances of the word "at" 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 @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ignon

-0.18

este

-0.18

oman

-0.15

istrovstvÃŃ

-0.15

iji

-0.15

odia

-0.15

ero

-0.14

ymm

-0.14

asts

-0.14

 duct

-0.14

POSITIVE LOGITS

azzo

0.17

ndl

0.16

ripp

0.15

 Telescope

0.14

èĩ

0.14

apon

0.14

_LVL

0.14

 KrÃ¡l

0.14

iazza

0.14

celed

0.14

Activations Density 0.019%

instances of the word "at" in various contexts

No Comments

No Known Activations