INDEX

Explanations

references to research, particularly in academic contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 researched

-0.24

 research

-0.24

 Research

-0.23

Research

-0.21

 researching

-0.20

research

-0.20

 recherche

-0.20

 researcher

-0.19

çłĶç©¶

-0.18

 ricerca

-0.18

POSITIVE LOGITS

es

0.43

Gate

0.25

 into

0.23

er

0.22

 conducted

0.21

ES

0.21

gate

0.21

esin

0.20

 Triangle

0.20

 findings

0.20

Activations Density 0.054%

references to research, particularly in academic contexts

No Comments

No Known Activations