INDEX

Explanations

references to educational initiatives and collaborative efforts aimed at improvement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

hurst

-0.17

Ã¸y

-0.15

linkplain

-0.15

ÐµÐ²Ð¸

-0.15

ableObject

-0.15

vvm

-0.14

versible

-0.14

 uptake

-0.14

woods

-0.14

edBy

-0.14

POSITIVE LOGITS

 Free

0.23

Gri

0.21

Free

0.20

FS

0.18

Kar

0.18

_Free

0.18

 Kimberly

0.18

Nama

0.18

Wit

0.18

-Free

0.17

Activations Density 0.040%

references to educational initiatives and collaborative efforts aimed at improvement

No Comments

No Known Activations