INDEX

Explanations

references to geographical locations and their characteristics

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

á»Ĩ

-0.16

ãĤĦãģ£ãģ¦

-0.16

ëŀ

-0.16

 váºŃy

-0.14

å±Ĭ

-0.13

å¢Ĺ

-0.13

adge

-0.13

istrovstvÃŃ

-0.13

 **)&

-0.13

^K

-0.13

POSITIVE LOGITS

 means

0.62

 mean

0.56

 Means

0.54

means

0.53

 meant

0.47

Means

0.46

Mean

0.44

 Mean

0.43

mean

0.43

-mean

0.40

Activations Density 0.298%

references to geographical locations and their characteristics

No Comments

No Known Activations