INDEX

Explanations

instances of the word "published" in the text

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

ices

-0.15

 Miami

-0.14

orton

-0.14

 mane

-0.13

bul

-0.13

 Schwartz

-0.13

ÄįÃŃ

-0.13

 Dirty

-0.13

owie

-0.13

emb

-0.13

POSITIVE LOGITS

ctxt

0.15

abra

0.15

ikit

0.15

à¸§à¸£à¸£

0.15

veloper

0.15

igy

0.15

 Inflate

0.14

USS

0.14

à¸µà¸¢

0.14

Activations Density 0.013%

instances of the word "published" in the text

No Comments

No Known Activations