INDEX

Explanations

geospatial information and descriptions related to landscapes 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 @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 Hills

-0.23

 hills

-0.21

ä¸ĺ

-0.21

hil

-0.19

 Hill

-0.19

 hill

-0.17

ship

-0.16

 Guild

-0.15

hill

-0.15

 Carm

-0.15

POSITIVE LOGITS

 glacier

0.23

 glaciers

0.23

hut

0.21

 snow

0.21

al

0.20

 Matter

0.20

 Refuge

0.19

kees

0.19

snow

0.19

 Alpine

0.18

Activations Density 0.051%

geospatial information and descriptions related to landscapes and locations

No Comments

No Known Activations