INDEX

Explanations

percentage-related terms indicating success or guarantees

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ello

-0.18

je

-0.17

-0.16

ule

-0.16

oje

-0.15

ole

-0.15

iola

-0.14

 coke

-0.14

Van

-0.14

egg

-0.14

POSITIVE LOGITS

?p

0.15

 Babe

0.15

ÙĨÚ¯ÛĮ

0.15

 Drill

0.15

portun

0.15

ÑĸÐ½ÑĮ

0.15

auer

0.15

ÑĦÐ¸ÑĨÐ¸

0.15

	throws

0.14

á»ĭch

0.14

Activations Density 0.020%

percentage-related terms indicating success or guarantees

No Comments

No Known Activations