INDEX

Explanations

place names and geographic features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

Î¿Îº

-0.14

jan

-0.13

":[{↵

-0.13

secs

-0.13

WARD

-0.13

 Castillo

-0.13

 Mell

-0.13

.Cart

-0.13

Âª

-0.13

_slices

-0.13

POSITIVE LOGITS

 Creek

0.45

Bay

0.33

 Hill

0.33

Gul

0.31

 creek

0.31

Gap

0.31

 Point

0.30

 Springs

0.30

 Mountain

0.29

 Bend

0.27

Activations Density 0.266%

place names and geographic features

No Comments

No Known Activations