INDEX

Explanations

phrases related to various directions or choices in context

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

uvo

-0.16

udic

-0.15

elly

-0.15

pons

-0.15

woord

-0.14

lah

-0.14

AndView

-0.14

ãģªãģ®

-0.14

urtle

-0.14

ÑģÑĮ

-0.14

POSITIVE LOGITS

 direction

0.67

 directions

0.59

direction

0.56

æĸ¹åĲĳ

0.56

 Direction

0.50

Direction

0.46

-direction

0.45

 Directions

0.45

_direction

0.42

.direction

0.41

Activations Density 0.163%

phrases related to various directions or choices in context

No Comments

No Known Activations