INDEX

Explanations

terms related to trade secrets and confidentiality

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

 Burke

-0.16

urb

-0.16

ware

-0.16

-0.15

Ware

-0.15

Nag

-0.15

ability

-0.14

 based

-0.14

 tender

-0.14

POSITIVE LOGITS

lez

0.16

issan

0.15

slu

0.15

hiba

0.15

ë§ī

0.14

ulace

0.14

ÑĤÑĢÐ¾

0.14

ä¾

0.14

orro

0.14

Ð¾Ð¼ÐµÐ½

0.14

Activations Density 0.020%

terms related to trade secrets and confidentiality

No Comments

No Known Activations