INDEX

Explanations

references to significant historical events or anniversaries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÙĩØ§Ø±

-0.18

elor

-0.16

zsche

-0.16

Åº

-0.15

.opens

-0.15

 Cocoa

-0.14

okit

-0.14

ÎµÎ»

-0.14

BERS

-0.14

uD

-0.14

POSITIVE LOGITS

 Jordan

0.54

Jordan

0.47

 Petra

0.34

 Jord

0.32

 Hash

0.25

ordan

0.25

 Dead

0.23

 Salt

0.23

 Utah

0.22

Am

0.21

Activations Density 0.013%

references to significant historical events or anniversaries

No Comments

No Known Activations