INDEX

Explanations

references to different sections within a document or text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ous

-0.18

oyo

-0.16

.ObjectMeta

-0.15

 fung

-0.14

ÏĤ

-0.14

veau

-0.14

Broad

-0.14

ollen

-0.14

inker

-0.14

efon

-0.14

POSITIVE LOGITS

naires

0.28

naire

0.26

ally

0.24

ately

0.19

nement

0.18

ality

0.18

alist

0.17

ALLY

0.16

als

0.16

nal

0.16

Activations Density 0.038%

references to different sections within a document or text

No Comments

No Known Activations