INDEX

Explanations

addresses and place names

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

-0.17

ifik

-0.15

 Hubb

-0.15

erties

-0.15

ãĥĩãĤ£ãĤ¢

-0.14

slash

-0.14

Zug

-0.14

 Museum

-0.14

Unmount

-0.14

POSITIVE LOGITS

 Trem

0.20

anych

0.19

 Label

0.18

 Cardinal

0.17

 Siemens

0.17

 MacDonald

0.16

ighet

0.16

 Forget

0.16

osit

0.15

 Champagne

0.15

Activations Density 0.084%

addresses and place names

No Comments

No Known Activations