INDEX

Explanations

expressions related to varying opinions or agreements

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

IPP

-0.16

ippet

-0.15

ahkan

-0.15

imson

-0.14

ulle

-0.14

abad

-0.14

aea

-0.14

 gezocht

-0.14

 Clerk

-0.14

 Spinner

-0.14

POSITIVE LOGITS

 others

0.18

 ones

0.16

 certain

0.16

 Sterling

0.15

ugg

0.15

dem

0.15

 primary

0.15

arian

0.14

 Keller

0.14

/command

0.14

Activations Density 0.138%

expressions related to varying opinions or agreements

No Comments

No Known Activations