INDEX

Explanations

references to automotive or mechanical professions and education

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

 doktor

-0.17

 doctoral

-0.17

/umd

-0.16

 faculty

-0.16

 professors

-0.16

æķĻæİĪ

-0.16

stalk

-0.16

 Consultants

-0.15

aylight

-0.15

 doctor

-0.15

POSITIVE LOGITS

 technician

0.21

 Technician

0.19

 helpers

0.19

 technicians

0.18

 Helper

0.17

 helper

0.17

 Train

0.17

 train

0.17

 lineman

0.17

ãĤ¢ãĥ«ãĥĲ

0.16

Activations Density 0.247%

references to automotive or mechanical professions and education

No Comments

No Known Activations