INDEX

Explanations

phrases indicating ownership or possession

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

-0.16

èº«ä¸Ĭ

-0.15

zn

-0.15

esel

-0.14

ets

-0.14

 einzel

-0.14

zac

-0.14

lea

-0.13

enga

-0.13

eden

-0.13

POSITIVE LOGITS

 larger

0.25

 wider

0.23

 overall

0.21

overall

0.20

larg

0.20

 Larger

0.20

 ongoing

0.20

 suite

0.20

 effort

0.19

 broader

0.19

Activations Density 0.041%

phrases indicating ownership or possession

No Comments

No Known Activations