INDEX

Explanations

structured data or references to experiments and methodologies

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

phon

-0.16

.scalar

-0.14

.gdx

-0.14

 PROFILE

-0.14

emode

-0.13

cust

-0.13

PROFILE

-0.13

phem

-0.13

å¤¢

-0.13

 cust

-0.13

POSITIVE LOGITS

 proposal

0.25

 proposals

0.25

 detection

0.24

 Detection

0.23

 object

0.23

 detections

0.23

 detectors

0.23

 detector

0.22

 Proposal

0.22

Io

0.22

Activations Density 0.008%

structured data or references to experiments and methodologies

No Comments

No Known Activations