INDEX

Explanations

repeated suffixes, particularly those ending with "on" or "ons."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

owl

-0.15

-0.14

on

-0.14

pat

-0.14

ouch

-0.14

Jun

-0.14

anship

-0.13

Sibling

-0.13

atum

-0.13

rom

-0.13

POSITIVE LOGITS

ÐµÐºÑģÐ¸

0.17

aires

0.17

forge

0.16

naire

0.16

voir

0.15

exion

0.15

UPLE

0.15

.infinity

0.15

AMESPACE

0.15

Äįer

0.15

Activations Density 0.075%

repeated suffixes, particularly those ending with "on" or "ons."

No Comments

No Known Activations