INDEX

Explanations

concepts related to philosophical questions and analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

 physical

-0.16

Mr

-0.15

 pare

-0.15

 inst

-0.15

Gov

-0.14

arak

-0.14

ulace

-0.14

abi

-0.14

ello

-0.14

POSITIVE LOGITS

 pole

0.18

 disciplinary

0.16

pole

0.16

æŁ»

0.15

 Reception

0.14

/↵↵↵↵

0.14

 Pole

0.13

Ø¹ÙĦØ§Ùħ

0.13

continental

0.13

iscard

0.13

Activations Density 0.134%

concepts related to philosophical questions and analysis

No Comments

No Known Activations