INDEX

Explanations

references to travel and locations

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

Republic

-0.16

 Jama

-0.15

 Republic

-0.15

umo

-0.14

ulis

-0.14

 Hawaiian

-0.14

 Trinidad

-0.14

 FileType

-0.14

.DOM

-0.14

tic

-0.14

POSITIVE LOGITS

 Virgin

0.20

Virgin

0.19

 Tort

0.17

VI

0.16

 Schneider

0.16

evi

0.16

 Brewers

0.15

VI

0.15

St

0.15

ahl

0.14

Activations Density 0.024%

references to travel and locations

No Comments

No Known Activations