INDEX

Explanations

references to the concept of "bottom" related to physical objects or positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

/cms

-0.15

 Psychological

-0.15

upply

-0.14

toi

-0.14

Utf

-0.14

å·±

-0.14

 psychological

-0.14

 Rosenberg

-0.14

IENT

-0.13

ë²½

-0.13

POSITIVE LOGITS

 below

0.23

below

0.20

 bottom

0.20

ä¸ĭçļĦ

0.19

.Bottom

0.18

 beneath

0.18

bottom

0.17

 Below

0.17

-Level

0.17

 lowest

0.17

Activations Density 0.149%

references to the concept of "bottom" related to physical objects or positions

No Comments

No Known Activations