INDEX

Explanations

words and phrases that denote significant or impactful concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

ity

-0.37

theless

-0.36

istrator

-0.35

ation

-0.35

Ø©

-0.34

ï¸ı

-0.33

ìĦľ

-0.31

thing

-0.31

ive

-0.30

plier

-0.30

POSITIVE LOGITS

íļĮìĿĺ

0.18

ards

0.17

±Ð¾ÑĤ

0.16

Ä±mÄ±n

0.16

UNET

0.16

DbType

0.15

rovers

0.15

lesai

0.15

ROTO

0.15

Î½Î±ÏĤ

0.14

Activations Density 0.394%

words and phrases that denote significant or impactful concepts

No Comments

No Known Activations