INDEX

Explanations

terms and categories related to content organization and classification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ãĤ¤ãĥĦ

-0.18

agnost

-0.16

íĬ

-0.15

jing

-0.15

ella

-0.15

Ø§Ø·

-0.15

ucci

-0.14

ozor

-0.14

rovers

-0.14

osy

-0.14

POSITIVE LOGITS

 Bull

0.17

UTE

0.16

@gmail

0.15

ERO

0.15

ew

0.14

 Shank

0.14

iten

0.14

imen

0.14

umer

0.13

UTO

0.13

Activations Density 0.002%

terms and categories related to content organization and classification

No Comments

No Known Activations