INDEX

Explanations

expressions related to adding finishing touches or enhancements, particularly those that emphasize personal and aesthetic improvements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ActionTypes

-0.16

Subset

-0.14

itchens

-0.14

eneric

-0.14

Outlined

-0.14

itchen

-0.14

uforia

-0.14

Naz

-0.13

CTIONS

-0.13

POSITIVE LOGITS

 touch

0.69

 touches

0.68

 Touch

0.57

 TOUCH

0.55

touch

0.55

-touch

0.55

touches

0.52

Touch

0.50

 touched

0.49

 finishing

0.47

Activations Density 0.083%

expressions related to adding finishing touches or enhancements, particularly those that emphasize personal and aesthetic improvements

No Comments

No Known Activations